Coral reefs are among the most diverse ecosystems in the world, harboring over one quarter of all known marine species. However, reefs are under tremendous stress from a variety of sources at all scales, including Malthusian overfishing and mass coral bleaching associated with global climate change. Traditional site-based management has proved to be insufficient to protect reefs against such extensive and devastating threats. Most scientists agree, however, that well-designed networks of no-take reserves within Marine Protected Areas (MPAs) can provide stability against ecological disasters and aid in long-term fisheries management. Furthermore, these reserves can serve as genetic reservoirs of biodiversity and sources of recruits for re-colonizing damaged areas. One of the key questions in designing such networks is to what extent the reserves are connected by larval dispersal, ensuring that populations are not isolated. This connectivity can be difficult to assess due to variable current patterns and recruitment dynamics. However, data from an existing long-term monitoring and mapping project makes it feasible to model these large-scale complex processes. One such project is the University of Hawaii's Coral Reef Assessment and Monitoring Program (CRAMP). I propose to analyze CRAMP's data on fish and coral population structure to identify potential larval sources, evaluate existing oceanographic data, and conduct current studies to estimate larval dispersal. This combination of modeling and experimental approaches will provide critical information for the design and management of networks of interconnected marine reserves, both in Hawaii and in other marine ecoregions.