FROM THE VANTAGE OF AN OPEN-COCKPIT AIRPLANE, the colors of the water below shift from light green to aqua blue to rich cobalt. These color changes signify the different patches of habitat that lie beneath the surface: sea grass beds, shallow reefs, halos of sand, and deep channels. From 300 m (1,000 ft) above, this habitat mosaic hints at the connectedness of the coral reef ecosystem. Habitat patches of varying shapes and sizes are linked to one another by movement of fish, nutrients, and energy flows, creating patterns of connectivity across spatial scales from the reef patch to the barrier reef. Because the elements of coral reef ecosystems are interconnected, managing them as a whole, rather than as a series of disconnected parts, makes sense. Landscape ecology is ideally suited for addressing real-world problems, from designing migration corridors for elephants in Africa to creating networks of protected areas for giant otters in the Peruvian rain forest. By using a “seascape” perspective, scientists and managers can address concerns at a scale appropriate to resource management needs in coral reef ecosystems (Forman 1995).

Such a bird’s-eye or seascape perspective will be critical for designing marine protected areas (MPAs) (fig. 1), which have grown increasingly popular as ecosystem-based tools for improving fisheries management and protecting biodiversity. Effectiveness is contingent on understanding key ecological patterns and processes at appropriate spatial scales, and may depend upon maintaining critical linkages among essential habitat patches to conserve reef fish communities. Unfortunately, how the design of any particular MPA will influence its effectiveness, especially in highly complex and dynamic coral reef ecosystems, is unclear. Coral reefs exist as heterogeneous ecosystems consisting of various habitat patches that can differ markedly in their spatial arrangement and composition. Evidence is increasing for the importance of maintaining habitat linkages that support key ecosystem processes, which in turn sustain viable reef fish populations. A failure to consider and maintain functional habitat linkages that affect important ecological processes in the design of protected areas is likely to result in undesirable changes in community structure and possibly the loss of key species. Consequently, while reef habitat may be essential to reef fish production, different MPA designs—size, shape, and proximity to habitat types—may produce different results. For example, movement of fishes may be more likely outside the boundaries of MPAs given certain arrangements of habitat types (see Chapman and Kramer 2000). Thus, a key to future success in protecting reef habitat and fisheries is quantifying how alternative design options will impact a desired outcome. Simply setting aside areas based on jurisdictional or political motivations may not be sufficient.