Anisotropic shrinkage of mineral and organic soils and its impact on soil hydraulic properties

• Published in: Soil & tillage research 2012-09, Vol.125, p.96-104
• Main Authors: Gebhardt, Stephan, Fleige, Heiner, Horn, Rainer
• Format: Article
• Language: English
• Subjects: Agronomy. Soil science and plant productions; Biological and medical sciences; COLE value; Fundamental and applied biological sciences. Psychology; Physical properties; Physics, chemistry, biochemistry and biology of agricultural and forest soils; Shrinkage curve; Shrinkage geometry; Soil science; Unsaturated hydraulic conductivity; Water and solute dynamics; Water retention curve
• ISSN: 0167-1987; 1879-3444
• DOI: 10.1016/j.still.2012.06.017

► The actual anisotropic shrinkage of organic and mineral soils was studied. ► The observed shrinkage-anisotropy is desiccation-dependent. ► Considering shrinkage affects the θ(Ψ)- and k(Ψ)-relationship. ► For the determination of the θ(Ψ)- and k(Ψ)-relationship shrinkage must be observed. ► The observed shrinkage and shrinkage anisotropy, respectively and its consequences on hydraulic functions affect both the studied mineral and organic soils. The shrinkage behavior of soils does not only affect the aggregate formation but also it alters pore size distribution and pore functions like water and gas permeability. In order to quantify this shrinkage effect both with respect to the intensity and orientation (isotropy or anisotropy) undisturbed organic and mineral soil samples (sedge peat, peat-clay, half-bog and glacial clay) in Northern Germany were analyzed. In addition to the registration of the vertical volume change during drying by means of height change measurements assuming shrinkage isotropy, digital photography and image analysis were used to consider horizontal shrinkage components including crack formation in order to calculate the 3 dimensional shrinkage geometry (=kind of anisotropy). The organic substrates show the highest shrinkage potential according to the COLE (coefficient of linear extensibility) with values between 0.26 and 0.43. Maximum volume decreases from saturation to complete desiccation of up to 66vol.% (peat-clay) were calculated. In contrast to the mineral soils, the investigated sedge peat did not exhibit residual or zero shrinkage. Moreover, the soil volume decrease during desiccation partly even exceeded the water loss in case of the organic substrates (sedge peat, peat-clay) which can be attributed to mineralization. Shrinkage-dependent volume changes at initial desiccation were determined to be almost exclusively vertical (geometry factor rs=1). During further drying, however, shrinkage led to isotropy (rs=3) followed by a slight dominance of the horizontal shrinkage component. These results are true not only for shrinking mineral, but also for the tested organic soils. These results underline that a general assumption of a rigid pore system is hardly true. Moreover, the exclusive measurement of the shrinkage-induced soil height changes assuming shrinkage isotropy leads to a distinct overestimation of the actual shrinkage particularly at initial desiccation. Although the calculated COLE values were quite close to each other