Mechanisms of Soil Erosion in Subtropical Forests of China - Effects of Biodiversity, Species identity, Tree architecture and Spatial variability on Erosivity

Abstract

Soil erosion is a major threat to ecosystems and agricultural land worldwide. To overcome severe soil loss, aff orestation is used as a common tool. However, the mechanisms of soil erosion in forests are understood rarely up to now. There is still a knowledge gap to what extent biodiversity and tree species identity aff ect soil erosion in early successional forest stands, which tree architectural and leaf traits account for these eff ects and which of these traits are important for the spatial variability of soil erosion. Therefore, this thesis investigated the influence of tree species richness (as a measure of biodiversity) and tree species identity on rainfall erosivity (measured as throughfall kinetic energy; TKE). Furthermore, this thesis concentrated on the spatial variability of TKE. Importance and influence of five tree architectural and nine leaf traits on these TKE properties were evaluated. In addition, the influence of leaf litter diversity and soil meso- and macrofauna on initial soil erosion was investigated. The experiments were carried out in a young subtropical forest of southern China in the framework of the BEF-China (Biodiversity and Ecosystem Functioning) project. Tree species richness eff ects on TKE were found only at the local neighborhood scale while plot-level e ffects of tree species richness on TKE were not found. This eff ect was attributed to the young age of the forest plantation. Crown cover, canopy layering or tree heights have not yet fully developed and thus only e ffects at a local neighborhood scale can be seen. Neighborhood e ffects on TKE were due to larger crown areas and taller tree heights in more diverse neighborhoods thus increasing TKE. TKE was highly species-speci c. TKE below Choerospondias axillaris and Sapindus saponaria were higher and TKE below Schima superba was lower than the mean TKE of all other eight species. Species-speci c eff ects of TKE occurred due to diff erences in tree architecture and leaf traits. By far, leaf habit, leaf area and tree height were most important in inducing species-speci c TKE diff erences by changing rain drop velocity and drop size. Furthermore, TKE was spatially variable. Below the fi rst branch of a tree individual TKE was lowest due to low rain drop velocities and small drop sizes. In contrast, TKE was highest in the middle of four tree individuals due to a low interception by a low LAI resulting in higher throughfall amounts. In addition, this thesis provides a ranking of abiotic and biotic factors according to their importance for predicting TKE. Leaf area, leaf area index, throughfall and tree height were the most important variables. These findings emphasize the interplay between abiotic factors as well as tree architectural and leaf traits for a successful TKE prediction. Considering soil erosion management, the erosive potential of TKE in the experimental forest plantation can be mitigated by smaller leaf areas than 70 cm2, lower tree heights than 290 cm, lower crown base heights than 60 cm, smaller leaf area index than 1, more than 47 branches per tree individual and by using single tree species neighborhoods. Initial soil erosion (measured as sediment discharge) was not influenced by leaf litter diversity, but positively aff ected by the presence of soil meso- and macrofauna. This faunal eff ect arises mainly from arthropods slackening and processing the soil surface and only marginally from fauna taking part in the decomposition of leaves leading to less coverage. Nevertheless, leaf litter coverage highly negatively influenced the occurrence of initial erosion. Summarizing, biodiversity eff ects on soil erosion were neither present by investigating tree species richness at plot-level nor by investigating leaf litter diversity. However, a positive e ffect on TKE at the local neighborhood indicates that this can change with a full-grown and dense tree canopy with further succession of the forest. It can be concluded that in young successional forest stands tree architecture, leaf traits, species identity and abiotic factors are more important in influencing the erosive potential of rain than biodiversity.