Tropical forests cover just 7% of the Earth s surface, but store 25% of the global terrestrial carbon pool. Since they are so rich in carbon, net loss or uptake of carbon by tropical forests has important implications for atmospheric CO2 levels. Thus, tropical forests can speed up climate change by net emission of CO2 or slow it down by net sequestration. The rise in atmospheric CO2 level since the onset of the Industrial Revolution and the resulting climatic changes have certainly affected tropical forest dynamics. Understanding the impacts of these changes so far is crucial to predict future responses. Recent studies have shown that tropical forests have acted as carbon sinks over the last decades. However, it is unclear whether these forests have accumulated biomass already over the longer periods of time during which CO2 levels have risen and climate has changed. It is also unknown what has caused the observed biomass increase. The proposed study has the objective to detect, explain and predict long-term climate change effects on tropical tree dynamics. To this end, I will apply three techniques which are new to this field: tree ring analysis, stable isotope measurements and tree growth modelling. This is the first pan-tropical study that analyses centennial-scale tree growth. The study will be conducted at three sites (Bolivia, Cameroon, Thailand) and includes 15 tree species. I will test for gradual changes in long-term tree growth using tree ring data. Stable isotopes (C, O) will be measured to reconstruct past climate and detect responses to CO2 rise. The influence of changes in forest dynamics on tree growth will also be assessed. Finally, I will develop and apply physiological tree growth models and population models to unravel the complex interactive effects of climatic changes, CO2 rise and light climate on tree growth and population dynamics.