Ground-based data suggest tropical forest carbon sinks might soon become less effective at slowing the accumulation of CO2 in the atmosphere. However, Earth System Models fail to reproduce the observed trends of forest carbon sink saturation/decline, partly due to important gaps in the existing data underlying these models. The DAMOCO project intends to fill one of those most important data gaps concerning a tropical region of paramount importance: the Congo Basin. In this project, we are first collecting and/or valorizing data covering all the successional gradients (pioneer, regenerating and mature forest) of the most dominant forest types of the Congo Basin. By combining different but complementary scientific approaches, we produce multi-temporal records of forest dynamics: (i) sub-daily estimates of carbon and water fluxes collected with eddy covariance equipment on the first flux tower in central Africa, (ii, iii) decadal-scale changes in carbon balance and biodiversity through repeated tree and tree trait measurements from a network of permanent forest inventory plots, and finally, (iv) millennial-scale changes in biodiversity through the analysis of ancient charcoal from soil profiles located next to the plots. Secondly, we will combine this data to parametrize a state-of-the-art process-based Land Surface Model (ED2) for the central Congo Basin. Once the model is optimized, we will use it to upscale carbon and biodiversity dynamics for the entire central Congo Basin, apply it to test forest resilience to climate and land-use change in the region and simulate future carbon and biodiversity dynamics under climate and land-use change scenarios. Here, we present the early progress and results of the project. In particular, we focus on our first ancient charcoal data which suggest a gradual change in species composition throughout the Holocene, related to past long- and short-term droughts and fluctuations in population density. Furthermore, we provide an outline of how this paleodata will be integrated and combined with the other data sources in the ED2 Land Surface Model.