Tropical forests are critical for the global carbon cycle as they are responsible for a significant uptake of the carbon emissions and store the equivalent of about 40 years of today’s annual fossil-fuel emissions. However, they are threatened by increasing anthropogenic pressures which cause large-scale structural changes, the most important of which could be the increase of lianas (woody vines). As liana abundance negatively impacts the tropical carbon sink, and liana increase appears driven by climate change, lianas could enforce a climate feedback and be partly responsible for the global decline in the forest carbon sink. In the absence of reliable modelling tools, it is however impossible to quantify the importance of such a coupling and hence to forecast its future evolution. This is mainly because we lack critical data for parameterizing and validating lianas in vegetation models. This project aims to fill this important gap by observing key processes of the energy and water cycles of liana-present forest patches. These data must serve to calibrate and validate the radiative transfer and water competition modules of a vegetation model that includes lianas. Doing so, model simulations will highlight the competition factors between lianas and trees, and reveal how liana-induced impacts on the water and energy cycles affect the climate feedback. This project will contribute to better understand tropical forests, which is crucial from a conservation perspective.