THE IMPORTANCE OF SAMPLING SCALE IN ECOLOGY - KILOMETER-WIDE VARIATION IN CORAL-REEF COMMUNITIES

Observations along the north coast of Jamaica in 1994 suggested that a reas of relatively high coral cover and high coral diversity occurred adjacent to reefs that have been in decline since 1980. This study was carried out to quantify these observations, to determine whether simi lar variation occurs elsewhere in the Caribbean, and to draw attention to the significance of kilometer-wide variation in coral reef communi ty ecology. The fore reef (10 m depth) of Discovery Bay, Jamaica, had <3% coral cover, 12 species of scleractinians and >60% macroalgae cove r and appeared typical of a highly degraded Caribbean reef. However, n eighboring reefs (<9 km away) in the same reef zone and at the same de pth had up to 19 species of scleractinians, significantly higher coral cover (5 to 23%) and lower macroalgae cover (10 to 39%). Three reefs along the south coast of St. John, US Virgin Islands (10 m depth), whi ch are protected within a National park, displayed comparable kilomete r-wide variation in coral cover (3 to 33%) and macroalgae cover (22 to 33%). The abundance of juvenile corals also displayed similar levels of kilometer-wide variation within both islands; These results demonst rate that several measures of coral reef community structure are highl y variable over a spatial scale of kilometers; further study of these patterns may prove valuable in discerning the causes and consequences of coral demise. The existence of such large-scale variation emphasize s the importance of careful choice of sampling scale in the design and interpretation of monitoring programs in community ecology.