Abstract
Detailed information on long-term growth rates and ages of tropical rain forest trees is important to obtain a better understanding of the functioning of tropical rain forests. Nevertheless, little is known about long-term growth or ages of tropical forest trees, due to a supposed lack of annual tree rings in
... read more
most tropical tree species. Aim of this PhD thesis was to study the occurrence of annual tree rings in six tropical tree species in the Bolivian Amazon, and to use these rings to reconstruct historical growth patterns and determine ages. These data were also used to evaluate the potential of four timber species for obtaining a sustainable timber harvest.
All six tree species included in this study formed annual tree rings due to the occurrence of a distinct dry season, which induces cambial dormancy. Rainfall-growth relationships show a positive relation between tree growth and the amount of rainfall during certain periods of the year, indicating that rainfall plays a major role for tree growth. Ages of trees of 60 cm diameter differed three fold between and within species. The maximum observed age in the whole study was 427 years. Differences between trees in their temporal growth pattern towards the canopy ("canopy accession patterns") were studied by analyzing relative growth changes. This revealed that most trees do not reach the canopy by steady growth, but rather by an irregular pattern of growth spurts (‘releases’) and stand-stills (‘suppressions’), probably mostly caused by temporal variation in light. The four non-pioneer species for which these canopy accession patterns were analyzed showed differences in how they attained the canopy and in the length of periods of slow growth, suggesting differences in shade tolerance. Growth rates were positively autocorrelated among different trees, which means that a tree that has a high growth rate in one year relative to other trees is likely to perform next year also better than other trees. This autocorrelated growth was strong and lasted for long time periods. Incorporation of autocorrelated growth in simulations led to higher and more realistic variation in age estimates, emphasizing the importance of autocorrelated growth for population and growth studies. Four of the investigated species are exploited for their timber and for these species timber yields at the next harvest after 20 years (the minimum cutting cycle set by the Bolivian regulations) were estimated using the long-term growth data obtained by tree ring analysis. Results show low recuperation of initially harvested timber volume at the next cutting cycle (20-30%) and it will therefore be difficult to sustain timber yield for these species. Using tree ring data has several advantages over growth rates obtained from permanent sample plot data. First, the ages of trees are directly determined, not based on simulations. Second, the growth rates from ring data are representative for those trees that successfully reached the canopy and third, assessment and incorporation of variation among tree in long-term growth resulted in higher and more accurate estimates of timber yield. Tree ring analysis should be added to the toolbox of research methods for tropical forest management and has great potentials to advance our knowledge on forest dynamics and species coexistence.
show less
