资源环境科技发展态势分析平台

The global-scale degradation of coral reefs has reached a critical threshold wherein further declines threaten both ecological functionality and the persistence of reef structure. Geological records can provide valuable insights into the long-term controls on reef development that may be key to solving the modern coral-reef crisis. Our analyses of new and existing coral-reef cores from throughout the Florida Keys reef tract (FKRT) revealed significant spatial and temporal variability in reef development during the Holocene. Whereas maximum Holocene reef thickness in the Dry Tortugas was comparable to elsewhere in the western Atlantic, most of Florida's reefs had relatively thin accumulations of Holocene reef framework. During periods of active reef development, average reef accretion rates were similar throughout the FKRT at similar to 3 m/ky. The spatial variability in reef thickness was instead driven by differences in the duration of reef development. Reef accretion declined significantly from similar to 6,000 years ago to present, and by similar to 3,000 years ago, the majority of the FKRT was geologically senescent. Although sea level influenced the development of Florida's reefs, it was not the ultimate driver of reef demise. Instead, we demonstrate that the timing of reef senescence was modulated by subregional hydrographic variability, and hypothesize that climatic cooling was the ultimate cause of reef shutdown. The senescence of the FKRT left the ecosystem balanced at a delicate tipping point at which a veneer of living coral was the only barrier to reef erosion. Modern climate change and other anthropogenic disturbances have now pushed many reefs past that critical threshold and into a novel ecosystem state, in which reef structures built over millennia could soon be lost. The dominant role of climate in the development of the FKRT over timescales of decades to millennia highlights the potential vulnerability of both geological and ecological reef processes to anthropogenic climate change.