FISC - St. Petersburg
Rock Key and Sand Key Reefs: Rock Key and Sand Key Reefs are neighboring Holocene reefs located on the shelf edge west of Eastern Dry Rocks and southwest of Key West (Fig. 97C). The area is best known for the definitive multiple tracts of outlier reefs, referred to in the literature as the Sand Key outlier reefs. The term was coined to describe coral reefs that developed seaward of and are disconnected from a shelf margin by unfilled backreef troughs (Lidz et al., 1991). In line drawings traced from sparker profiles, Enos (1977, his Fig. 40) was the first to show that two such tracts existed in the Rock Key/Sand Key Reefs area. Three tracts were later documented seismically (Figs. 91B, 106A; Lidz et al., 1991). Four tracts were ultimately established photographically (Fig. 106B; Lidz et al., 2003; also see Benthic Ecosystems for Tile 7 and Tile 8).
A 6.09-m-long core (USGS-1) was drilled in the center of the largest outlier (Fig. 89, line 16a). Conventional uncorrected 14C dates on recovered corals indicated they were older than 38 ka (Lidz et al., 1991). Later dated with high-precision radiometric methods, corals in that core yielded ages of ~83.2 and 80.9 ka (Ludwig et al., 1996). A nearby seaward core (USGS-3) yielded an age of 8.2 ka (Fig. 106A; Ludwig et al., 1996). Pleistocene corals beneath those dated at 8.2 ka yielded an age of 82.7 ka (Toscano, 1996). Toscano (1996) drilled an additional core (SKOR2A) through the seaward base of a coral pinnacle landward of the USGS-1 core and obtained high-precision radiometric ages ranging from 106.5 to 81.6 ka in the Pleistocene section and 8.0 ka at the top. In a discussion of sea-level fluctuations during deep-sea oxygen-isotope Stage 5 (Fig. 37), Toscano and Lundberg (1999) correlated the Pleistocene ages with high sea-level stands during substages 5c (106 ka), 5b (94-90 ka), and 5a (88-80 ka). A core through the center of the pinnacle (USGS-2) yielded ages from 8.9 to 6.7 ka and a reworked coral clast with an age of 86.2 ka (Toscano and Lundberg, 1998).
Coring showed that the pinnacle was composed of cemented coral rubble. Scuba divers can see its composition in a 3-m-high outcrop on its seaward side, where the rubble is layered (E.A. Shinn, personal observation). It is believed that the pinnacle is the remnant of a rubble field that formed when sea level was at about the same elevation as the top of the pinnacle. At that time, the rubble field extended from the seaward Holocene reefs across the top of the outlier reef and was spilling into its backreef trough. With rising sea level and increasing wave energy, the uncemented parts of the field in the middle of the outlier reef were eroded down to the Pleistocene surface (Fig. 106A). Ages and locations of Holocene corals in the seaward reefs indicated coral growth had backstepped with rising sea level (i.e., younger corals grew upward and landward over older corals).
In a comparison of outlier-reef ages, cores were also obtained in a landward-oriented transect across the Carysfort outlier reef off north Key Largo (upper Keys; Figs. 97B, 107B; Toscano, 1996; Toscano and Lundberg, 1998). As on the Sand Key outlier reefs, Holocene coral accretions on the Carysfort outlier are thicker (6-7 m) on the landward crest than on the seaward forereef slope (2-3 m). Coral dates on the Carysfort outlier confirm at least one instance where Pleistocene corals backstepped ~0.25 km between 85.3 and 80.2 ka (Fig. 107B). Several other dates show that Holocene coral growth advanced seaward more than 0.5 km before backstepping toward the end of accrual (Toscano and Lundberg, 1998).
The differences in Holocene thickness and coral ages between the Carysfort and Sand Key outlier reefs reflect several factors relating to geographic location and a lower bedrock elevation to the southwest than northeast (see Bedrock Surface map).
The Sand Key Reef area is one of four shelf-margin sites for which an evolutionary cross-sectional model was developed based on seismic and coral-age data correlated with sea-level maxima (Figs. 37, 108A; Lidz, 2004). The models depict the origin, timing, development, and changes in surface landforms, i.e., the geomorphogeny. Methods for model construction are given in the discussion of Carysfort Reef (Fig. 36A). The other two sites modeled are The Elbow (Fig. 40A) and Pelican Shoal (Fig. 96A).
Comparison of the four models indicates that Pleistocene development of the shelf-edge reef was the same (upward and landward, partly filling a backreef trough) at The Elbow (north) and Pelican Shoal (south) and the same (upward, landward, and seaward) at Carysfort Reef (north) and Sand Key Reef (south). In each case, the Pleistocene component represents a backstepped reef-complex margin (Fig. 108A).
Landward growth is also suspected in the primary Sand Key outlier reef. The nearly 1-km-wide base of the largest Sand Key outlier (Fig. 108A) implies initial presence of an extensive reef flat, or gradual development of one over which corals later backstepped. Pleistocene coral dates from three outlier-reef cores confirm that backstepping took place at 81.6 and 80.9 ka (Fig. 106A; Toscano and Lundberg, 1998). Backstepped Holocene reefs on the shelf are well documented (e.g., Shinn, 1980a; Lidz et al., 1985).
Interpreting this type of margin in seismic profiles across an ancient platform can be problematic, because contacts between seismic coral facies would dip landward (Fig. 108B). Based on dip and absent other data indicating orientation relative to an ancient sea, evolution could be misinterpreted as seaward growth.