In this thesis, I have created and analysed three new datasets, providing information about leaf, tree and ecosystem responses to drought (Chapter 2), wood properties and traits in relation to drought sensitivity (Chapter 3) and above-ground growth responses to drought (Chapter 4 & 5) in the Amazon forest. Despite that no new data was added to the existing literature, I believe that the various presented analyses using previously published data have merit and contribute to the understanding of complex drought responses in humid neotropical forests.Overall, this thesis shows that the analysis of new datasets consisting of existing data and improved model representations can significantly improve our understanding of leaf, tree and ecosystem scale responses to drought in the Amazon forest. While I offer some first attempts to systematically analyse drought responses, to link wood anatomy to tree hydraulic architecture and drought sensitivity, and to upscale and model above-ground growth responses to drought in the Amazon, shortcomings and major uncertainties clearly remain. New field campaigns, improved data sharing, novel remote sensing products and new experimental work on the processes examined in this thesis could greatly benefit our understanding of drought responses in the Amazon and help to closely monitor the expected response of Amazon forest functioning to climate change.