Journal of Soils and Sediments

, Volume 10, Issue 3, pp 484–493

Shrinkage processes of a drained riparian peatland with subsidence morphology


DOI: 10.1007/s11368-009-0130-9

Cite this article as:
Gebhardt, S., Fleige, H. & Horn, R. J Soils Sediments (2010) 10: 484. doi:10.1007/s11368-009-0130-9



The drainage of a riparian peatland interspersed with organic and mineral substrates in Northern Germany has led to subsidence morphology due to mineralisation, initial settlement and shrinkage processes causing cultivation problems and yield declines. To explain the relevancy of shrinkage to changes of the surface morphology and soil properties, organic and mineral substrates were analysed with regard to (1) shrinkage potential (i.e. the maximum shrinkage), (2) shrinkage curve, (3) actual in situ shrinkage and (4) irreversible shrinkage proportions.

Materials and methods

The investigated substrates are sedge peat, peat-clay, half-bog and alluvial sand of Fibric Limnic Histosols (Drainic) and a glacial clay for comparison. The methods used to obtain the shrinkage curves include the consideration of crack formation due to desiccation, as digital photography and image analysis were used in addition to height change measurements. To characterise the actual in situ shrinkage, continuous records of the soil matric potential for a 4.5-year period at two different intensively drained Histosols (moderately and deeply drained: mean groundwater table 0.6 and 1.7 m under ground level) were accomplished in four depths (15, 40, 70, 100 cm).


The measured shrinkage potentials vary between 2% (alluvial sand) and 65% (peat-clay). The sedge peat and peat-clay are characterised by distinct higher shrinkage potentials compared to the glacial clay and show distinct proportional shrinkage. It has to be pointed out that the soil volume loss partly even exceeds the loss in water volume. The sedge peat does not reach residual or zero shrinkage but shows distinct structural shrinkage at initial desiccation. For the moderately drained site, matric potentials reached up to −60 kPa in summer in 15-cm depth, whilst capillary rise almost prevented desiccation of deeper zones. For the deeply drained site, summer desiccation reached deeper zones as well (matric potential up to −50 kPa in 40-cm depth) due to absent capillary rise. In the range of actually occurring maximum desiccation, the half-bog and peat-clay with high decomposed organic matter are particularly affected by shrinkage with volume decreases of 10–14%, in contrast to the fibred sedge peat with only 4%. When uniquely shrunken substrates will be re-wetted, distinct irreversible shrinkage proportions reaching between 18% and 31% subsequent to drying to a matric potential of −50 kPa and re-saturation were measured.


Particularly, the organic-rich substrates appeared as characteristically sensitive to soil shrinkage. Contrary to mineral substrates, they lack in solid particles, leading, along with high hydraulic stresses due to dominantly finer pores, to intense volume decreases after desiccation. These substrates therefore partly also lack in residual or zero shrinkage, and the soil volume loss can even exceed the water loss due to shrinkage. Accordingly, the alterations of soil properties, e.g. conductivities, are especially pronounced.


The calculated proportion of actual in situ shrinkage (0.5 to 1 dm) to total subsidence of the investigated peatland is rather low for the deeply drained site in comparison to initial settlement (2–3 dm) and long-term active mineralisation (e.g. 4 dm in 40 years, mineralisation rate 1 cm year−1) but accelerates the latter because of enhanced aeration due to crack formation. Irreversible shrinkage is one reason why re-wetted fen peats do not retrieve their initial volume as well as their initial physical and hydraulic properties. Thus, a re-wetting of drained fen peats can only preserve the present condition.

Recommendations and perspectives

To impede the further impairment of riparian peatlands, it is necessary to maintain a near-surface water table particularly throughout the summer months, when high desiccation occurs, leading to a spatially differentiated proportion of initial settlement, shrinkage and mineralisation, causing subsidence morphology. Our recommendation is a drainage depth of less than 60 cm to enable pasture or grassland land use and to minimise mineralisation as well as structure formation in the (sub)soil.


DrainageHistosolRiparian peatlandShrinkage curveShrinkage potential

Copyright information

© Springer-Verlag 2009

Authors and Affiliations

  1. 1.Institute for Plant Nutrition and Soil ScienceChristian-Albrechts-University KielKielGermany