Fig. 3. Carbonate-budget models showing how increases in Acropora palmata through restoration could improve the ability of reef-crest habitats at Buck Island Reef National Monument (BIRNM) to keep pace with future sea-level rise.

These results indicate that if A. palmata cover could be increased by ~30% (to total mean cover of 36 and 45% for the northern and southern reef crest, respectively), the reef could keep pace with Low to Intermediate-Low projections of relative sea-level (RSL) rise. Median projections (solid to dashed lines; shading represents two-standard-deviation confidence intervals) of future relative sea-level rise for St. Croix (relative to 2016; NOAA Tide Station # 9751401; ref. ³¹) to 2050 a and 2100 b are based on the five (Low to High) sea-level-rise scenarios evaluated in Sweet et al.³¹. The sea-level-rise projections (left) are compared with carbonate-budget-based projections of median reef-elevation change (solid black horizontal lines) and reef-accretion potential (right; boxplots) of BIRNM reef-crest habitats based on our surveys in 2016 (+0%) and under restoration scenarios of 5–30% increases in A. palmata cover. Boxes bound the first and third quartiles and whiskers represent 1.5× the inter-quartile ranges. White triangles represent a lower uncertainty on median projected elevation changes based on an estimate of potential erosion processes not accounted for in the carbonate budgets (see Methods). Note that these uncertainties are generally encompassed by the 1.5× inter-quartile range of the boxplots in the more optimistic restoration scenarios. Areas representing negative elevation change and reef-accretion potential are shaded in gray.