Quantitative and qualitative assessments of fish and benthic community structure are being used in the development of a long-term monitoring programme for the Exuma Cays Land and Sea Park (ECLSP). Combined with estimates of critical resource thresholds for coral reef habitats, this monitoring promises to contribute to more effective management of the ECLSP, which in turn will help minimize future anthropogenic impacts and improve the health and resilience of critical habitats. Fourteen sites, across four habitat types deep forereef (> 10 m), shallow forereef (< 10 m), fringing, and channel reefs, were selected based on relative potential impacts from several human activities such as fishing, diving, and coastal development. Species density and biomass data were collected for fish species observed during belt transect surveys. Relative abundance of commercial fish species and invasive lionfish was assessed using transect and timed roving diver surveys. Dominant benthic cover was estimated using point-intercept sampling within 1-m2 quadrats. Species lists of major taxa were also compiled for each quadrat. Preliminary data analyses of fish species abundances using Analysis of Similarity (ANOSIM) and multidimensional scaling (MDS) revealed temporally stable fish communities that varied by habitat type. The greatest differences in fish community structure were between two of the forereef sites outside of the Park and the rest of the forereef sites. ANOSIM and MDS analyses show that benthic community structure and biodiversity composition varied significantly across sites and reef types. Deeper and shallow forereefs differed significantly, and both were highly distinct from fringing and channel reefs, which were not significantly different. Forereef sites were dominated by macroalgae; fringing and channel reefs were dominated by combinations of corals, sponges, and turf algae. Analyses detected no obvious impacts from diving and development pressures on reefs, but statistical power remains poor given the number of sites and the duration of the sampling period.