Southeastern Section - 57th Annual Meeting (10–11 April 2008)

Paper No. 3
Presentation Time: 8:40 AM


VANWALLEGHEM, Tom1, POESEN, Jean2 and DECKERS, Seppe2, (1)Agronomy, University of Cordoba, Avd. Menendez Pidal s/n, Cordoba, 14010, (2)Leuven, 3000,

The wide variability of soil profiles across landscapes is known since the earliest soil observations and forms the basis for the catena concept. Soil horizon depth is a critical variable in a number of environmental problems, ranging from hydrological modelling to landscape reconstruction. Despite its importance, almost no information exists on how the depth of soil horizons varies in undisturbed, natural landscapes. The reason for this is probably mainly due to the fact that sites that have never been influenced by soil erosion are rare.

This study aims at documenting and explaining this variation in soil horizon depth in function of terrain parameters. For the particular objectives of this study, focus will be on loess-derived soils (Luvisols and Albeluvisols).

This analysis is done in a unique natural archive in the Belgian loess belt, the Meerdaal Forest. In total 399 augerings of up to 8,7 m deep were made in an area of 1329 ha. The variability of 5 different soils horizons was evaluated. The most suitable marker for landscape reconstruction, because of its depth to the surface, is the border between decalcified and the original, calcareous parent material. This border showed important variation: between 1.14 and 5.10 m. A clear relation could be shown with landcape position and slope. Decalcification depth was significantly highest in valley bottoms, intermediate on plateau positions and lowest on slopes. With increasing slope gradient, decalcification depth was significantly lower. Also wetness index appeared to be significantly related with the depth of the decalcification border.

This study clearly demonstrates the invalidity of the traditional approach in landscape reconstruction, whereby past soil surfaces are reconstructed assuming a constant marker horizon. It provides important data for constructing soil-landscape models that dynamically predict soil horizons in function of landscape geomorphology.