Water repellency in sandy luvisols under different forest transformation stages in northeast Germany

Soil water repellency can adversely affect soil hydrological properties, reduce infiltration capacity and induce preferential flow, surface runoff and erosion. Although primarily described from semiarid and Mediterranean climates, it is also a concern in forest sites under the humid climatic conditions of central Europe. The objective of this study was to characterize the water repellency on geologically comparable forest sites with different stand ages and tree species in terms of the effects of forest transformation upon soil physical proper-ties. The investigated forest site called "Kahlenberg" (50 km northeast of Berlin) contains populations of Pinus sylvestris and Fagus sylvatica of different ages and proportions of single tree species. For the characterization of water repellency, the water drop penetration time (WDPT) test, for determining the persistence of water repellency, and the ethanol percentage (EP) test, which measures the severity or degree of water repellency, were carried out using soil samples from four forest plots and different soil depths (0-160 cm). "Potential" water repellencies Were determined after 3-day oven-drying at 45 degreesC. For two plots, the "actual" water repellency with field moist soil samples was also measured using the WDPT test. The WDPT test revealed a significant Proportion of severely and extremely hydrophobic samples in the upper 10 ern of the soil profile for all plots, whereas the persistence of repellency decreased with increasing soil depth. Soil organic matter contents and measured WDPT gave a positive linear correlation with r = 0.73. Maximum measured ethanol percentages were 30% and restricted to the upper 5 cm of the soil profiles. The correlation of the organic matter content with EP was r = 0.83. For all plots, the EP exhibited a shallower depth distribution than the WDPT. The two plots with pure pine and beech stands exhibited relatively low water repellencies, compared to the plots with mixed tree populations. This corresponds to a higher proportion of mor-type humus and a greater thickness of the humose topsoil at mixed stands compared to the pure pine and beech sites. (C) 2002 Elsevier Science B.V All rights reserved.