Tropical forests harbor a greater diversity of organisms than any other ecosystem and play a crucial role in the global carbon cycle – accounting for more than a third of terrestrial carbon fixation and half of aboveground biomass carbon storage. However, in part because of their biodiversity, the ecology of tropical forests remains relatively poorly understood in comparison with temperate ecosystems. In the last few years, research combining experiments and observations with mechanistic modeling has emerged as a powerful approach to glean new insights into these complex ecosystems. Mechanistic models parameterized and evaluated with detailed empirical data have provided frameworks to test alternative hypotheses, disentangle the roles of multiple drivers, and investigate the importance of plant functional variation and diversity itself to ecosystem processes. Data-model integration studies in tropical forests have targeted multiple processes and levels of organization, from plant physiology to individual growth to population demography to community composition to ecosystem productivity and nutrient cycling. This session presents a variety of studies utilizing different models and different types of empirical data at different study sites to address multiple fundamental and applied questions in tropical forests. Studies here explore the roles of interspecific variation in plant traits related to shade-tolerance, height and crown allometries, leaf phenology, hydraulic architecture, and/or rooting depths, in combination with environmental variation in climate, disturbance history, and soils, and its consequences for individual tree growth, species coexistence, woody productivity, and forest biomass, among others. The outcomes of these research efforts are not only an improved understanding of tropical forests today, but also an improved ability to predict their responses to ongoing anthropogenic global change. This itself is also a major accomplishment, as tropical forests have been one of the two largest sources of uncertainty in projecting future global carbon budgets, because of critical uncertainties in their responses to increasing atmospheric carbon dioxide, temperature, and drought stress. This session will be of interest not only to empiricists and modelers working in tropical forests, but also to any ecologists interested in learning more about how data-model integration can enable new insights into ecosystem function, even in the face of tremendous diversity and complexity.