RT Dissertation/Thesis T1 Tree growth and tree regeneration in two East African rain forests as related to the abiotic environment after human disturbance A1 Gliniars,Robert WP 2011/02/10 AB This study deals with the stem growth and seedling regeneration of different native tree species in two East African rainforests influenced by human disturbance in Kenya (Kakamega Forest) and Uganda (Budongo Forest), also considering spatially and temporally variable environmental influences. In the lower montane rainforest (1500 to 1700 m a.s.l.) Kakamega Forest (KF) surveys were conducted on trees ≥ 5 cm diameter at breast height (DBH) on an overall area of 2.08 ha (1 ha plot and 27 plots x 400 m2, 2007 individuals, equalling 965 ha-1), 3 inventories in 7 years). At the same time 91 tree species from 39 families were identified. In addition, in subplots within the permanent plots, 8441 saplings and seedlings ≤ 5m DBH distributed among 59 tree species from 29 families were recorded (on 432 m2, over 6 years). In the lowland rainforest (1000 to 1100 m a.s.l.) Budongo Forest (BF) a total of 1010 trees (1403 ha-1), distributed among 91 tree species from 30 families, were recorded in surveys on an overall area of 0.72 ha (18 plots ā 400 m2, 2 inventories in 5 years). During the observation period saplings and seedlings (on 72 m2, 708 individuals, in 3 years) belonging to 38 tree species from 18 families were recorded. The study areas were situated in forest areas characterised by different successional stages. These stages have emerged as a result of differing degrees of human disturbance for commercial purposes in the recent past. As this research was part of a joint project, highly resolved data on factors such as soil values, climate values, canopy characteristics and disturbance indices e.g. the Commercial Disturbance Index (CDI) was also available for a few permanent plots. It was established that the level of species richness and species diversity is low in areas with either no or very high levels of past commercial human disturbance (e.g. after clear-cutting), whereas areas with intermediate disturbance were seen to be richer or more diverse in species. The distribution of β-diversity also indicated a disturbance gradient. The degree of human disturbance was as well reflected in the successional stages within the structure of the forest areas. The mean stem volume of trees in BF increased as the disturbance index decreased, in KF however the highest stem volume occurred in areas with intermediate CDI. The tree populations in areas with a low level of disturbance in BF were characterised by high volumes together with lower stem numbers per plot and were more stable, as manifested in the lower turnover. The openness of the canopy varied in the differently disturbed areas and had a positive effect on seedling density in BF, which in turn underlines the importance of light for the regeneration of seedlings. In order to test the regeneration capacity of individual tree species for the most part in the intermediate stages of succession in both forests, the stem increment of 1198 trees (≥ 10 cm DBH) was measured monthly using dendrometer bands. At the same time the phenology of leaves, flowers and fruits of these trees was monitored over a period of three (in BF) and five years (in KF). In addition regeneration and establishment dynamics were analysed by counting, identifying and measuring seedlings (≤ 100 cm high) in the subplots every three months. The tree species could then be classified and allotted to their position along the pioneer-climax continuum. The early successional pioneer species were characterised by rapid growth and higher mortality rates, had no to few seedlings in the understorey and were shade intolerant. The late successional climax species however grew more slowly, had a low mortality rate, were well-established as seedlings in the understorey and shade tolerant. Most species possessed either characteristics somewhere in between these extremes or both characteristics of early and late successional species and were thus classified as intermediate. The annual stem increment varied considerably between tree species ranging from 0.6 to 2.5 mm in BF and from 0.6 to 4.4 mm in KF, and the relative diameter growth ranged from 0.3 to 1% in BF and from 0.25 to 1.58% in KF. The variability in the growth rates within tree species was even higher than that between species. Moreover the annual variability in growth within a species was higher in thinner than in thicker individuals. Individual trees however grew at a relatively constant rate. Beside the inherent growth potential, monthly changes in the climatic conditions also had an influence on the growth pattern of many of the tree species in KF and on all of the species in the more strongly seasonal BF. During the warm dry season the growth rate decreased and then increased again when the rainy season set in. This was confirmed by positive correlations between monthly growth rates and the sum of rainfall over two or three months respectively and also by strongly negative correlations between monthly growth and maximum temperature. The intensity of these relations varied according to species. The phenological features of most species were also dependent on the climate, e.g. leaf shedding or flowering in the dry season and fruiting in the rainy season. Seedling dynamics showed a tendency towards an increase in the populations in the rainy season and a decrease in the dry season depending on the species. The distribution of the tree parameters was not as clearly related to the spatial distribution of soil parameters. There were a few correlations between tree diversity and soil parameters in KF, positive correlations with acid-extracted Ca, C and N and negative with electrical conductivity. Furthermore significant correlations were established in KF between tree growth and Ca, C and N, as well as negative ones with Mn and Mg in the soil, whereby the intensity of the relationship was species specific. Seedling diversity also correlated positively with Ca, C and N, as well as negatively with Mn. The distribution and density of the seedlings was however more highly dependent on recurring pulses of newly recruited individuals at irregular intervals than on the abiotic environment. The study provides a first, precise insight into ecological regeneration processes of tree species populations and forest communities after human disturbance that have for the most part not yet been the subject of investigation, and can serve as a basis for forest management and forest models. K1 Tropischer Regenwald K1 Ostafrika K1 Pflanzenökologie K1 Keimling K1 Baumbestand K1 Boden-Pflanze-System K1 Tropenklima PP Hohenheim PB Kommunikations-, Informations- und Medienzentrum der Universität Hohenheim UL http://opus.uni-hohenheim.de/volltexte/2011/545