Partitioning of soil water among canopy trees during a soil desiccation period in a temperate mixed forest
- 1Department of Tropical Silviculture and Forest Ecology, Burckhardt Institute, Faculty of Forest Sciences and Forest Ecology, University of Göttingen, 37077 Göttingen, Germany
- 2School of Environment, The University of Auckland, Auckland 1142, New Zealand
Abstract. Complementary resource use is considered an important mechanism in the study of biodiversity effects. Here we explore how species identity, species mixture and tree size influence the vertical partitioning of soil water among canopy trees during a soil desiccation period. In the Hainich Forest, Germany, the species Fagus sylvatica, Tilia sp. and Fraxinus excelsior were studied in single- and three-species mixed clusters, each consisting of three co-dominant trees situated within a larger mixed forest stand. Vertical soil water uptake depth was assessed by analyzing the hydrogen stable isotope composition (deuterium, δD) of water from depth intervals throughout the soil profile and in tree xylem water. For single species clusters, a mixing model suggested that Fagus distinctively drew water from soil depths of 0.3–0.5 m, Tilia from 0.3–0.5 m and 0.5–0.7 m and Fraxinus mainly used water from 0.5–0.7 m. In mixed clusters, the uptake patterns of Fagus and Tilia were similar to those of the single-species clusters (mainly uptake form 0.3–0.5 m), but Fraxinus showed a different uptake pattern. Fraxinus in mixture had a somewhat homogenously distributed uptake over the soil depths 0.2–0.7 m. For single species clusters, there was no correlation between main soil water uptake depth and tree diameter, irrespective of variations in tree size. In contrast, for mixed clusters there was a significant decrease in the main uptake depth with increasing tree size (P<0.001, R2adj = 0.73), irrespective of species mix. In consequence, soil water partitioning was strongest where species were mixed and tree size varied. We further analyzed whether single and mixed-species clusters differed in the level of water uptake, e.g. due to complementarity, but our soil water budgeting did not indicate any such differences. A possible explanation might be that the volume of water used is predominantly governed by properties at the stand level, such as aerodynamic roughness, rather than by processes acting at the meter scale between neighbouring trees. With respect to application, we assume that the upcoming close-to-nature forestry approach for the area, which fosters mixed stands of heterogonous diameters, may result in enhanced complementarity in soil water uptake among canopy trees.