The first bimonthly resolved, paired δ18O and Sr/Ca time series from the slow-growing, tropical western Atlantic coral, Siderastrea siderea, from the Dry Tortugas, Florida has been generated and used to document that robust proxy climate records can be produced from this heretofore underutilized massive coral. The coral time series contains a 20-year long calibration window (1973-1992) for both δ18O and Sr/Ca and a 73-year long verification window (1900-1972) for Sr/Ca. These time series permits both the quantification of the relationship between coral δ18O-SST and Sr/Ca-SST using an augmented, 1° x 1° gridded SST record and the assessment of the stability of the proxy relationships over time. Both coral geochemical records are highly correlated with the augmented instrumental SST record through the calibration period and Sr/Ca remains highly correlated through the verification period both at the bimonthly (r = -0.97) and annual average level (r = -0.72). Additionally, both coral δ18O and Sr/Ca are highly reproducible within the same core, and Sr/Ca exhibits no extension-related vital effects. Sr/Ca-SST anomalies are also significantly correlated to the augmented SST anomalies, despite the removal of the serial autocorrelation. The skill of this proxy demonstrates its potential as a continuously growing, long-lived recorder of climate variability for the tropical Atlantic and Intra-American Seas. The relatively slow extension rate of the coral (~5 mm yr-1 during the 20th century) also suggests the potential for long records of climate variability (~200 years) of the region to be extracted from even modest-sized colonies (~1 m in height). The results of this study are important because relatively few century-long, sub-annually resolved time series of climate variability from massive Atlantic corals have been published, despite the significance of the tropical Atlantic climate modes of variability.

Core samples were collected in September 1995 from live coral heads of Montastraea annularis at Bird Key reef in the Dry Tortugas, Florida (24 degrees 55 minutes N, 82 degrees 92 minutes W). Four 4 mm-thick coral slabs from two cores were continuously sampled at 12 samples per year (0.025 cm per sample for Core 31, 0.023 cm per sample for Core 35). Visual inspection of X-radiographs indicates an average skeletal extension rate of about 3 mm per year in Bird Key corals. The goal of this study was to perform a replication test in Montastraea annularis by using elemental and stable isotopes from four coral slabs from two different coral heads to address the following questions: 1) how well do geochemical signals replicate within a single coral head, 2) how well do geochemical signals replicate from two different cores from the same coral head, 3) how well do geochemical signals replicate from two coral heads from the same general area, and 4) do growth effects influence the geochemistry of slow-growing corals at the Dry Tortugas? Geochemical variations versus depth and time of all coral records show strong seasonal cyclicity. Variations in d18O in the suite of Bird Key coral records replicate the best; d13C and Sr/Ca variations replicate less well. For example, differences in the mean Sr/Ca record from two different coral heads are large (0.179 mmol/mol for BK31B-BK35CC; 0.196 mmol/mol for BK31C-BK35CC; ~4 degrees C) and nearly 4 times greater than analytical precision. Therefore, caution must be exercised in interpreting Sr/Ca-SST records in Montastraea annularis. Mean differences in coral d18O for all records, on the other hand, are within analytical precision and translate to temperature differences of less than 0.5 degrees C. Robust d18O values among cores that co-vary with a significant level of agreement further point to this proxy being more reliable than Sr/Ca. Because of its skeletal complexity, drilling difficulty, and large bio-geological error for Sr/Ca, Montastraea annularis seems poorly suited for coral-based Sr/Ca-SST studies. However, the species must be studied to understand tropical Atlantic interannual-decadal scale variability, so further assessment is warranted.