Coastal ecosystems dominated by familiar components such algal and seagrass beds provide key ecosystem functions such as organic carbon production. However, they are facing various human-induced stresses, including local water pollution, coastal development and global climate change. The situation is serious and urgent especially in Asia – Western Pacific region where coastal biodiversity can be highest in the world, but is also the least studied. Seagrass beds are suitable for broad-scale comparisons of biodiversity and ecosystem functions because they occur worldwide from tropical to subarctic regions. I examined regional variation in biodiversity and abundance of seagrass communities from the Philippines to northern Japan to provide baseline data for future forecasting of coastal ecosystem changes with climate changes. Variables representing ecosystem functions were correlated with biodiversity measured at different spatial scales. However, the pattern of correlation varied greatly among regions, among different scales of observation, and among dependent variables in question. To elucidate complex causal mechanisms for the observed correlations, my collaborators and I are conducting a modular manipulative experiment. Tentative results suggest that the top-down/bottom-up effects on seagrass productivity vary substantially among different regions.