Journal of Conference Abstracts

Corals, Reefs, and Six Millennia of Holocene Climate

Ivan Gill (i_gill@rumac.uprm.edu)¹, Dennis Hubbard (dennis.hubbard@oberlin.edu)² & J. A. D. Dickson (jadd1@esc.cam.ac.uk)³

¹ Dept.of Geology, University of Puerto Rico, Mayaguez, PR 00681, USA

² Dept. of Geology, Oberlin College, Oberlin, OH, USA

³ Dept. of Earth Sciences, University of Cambridge, Downing St., Cambridge, CB2 3EQ, UK

Climatic reconstruction using corals has the potential to produce proxy records of sea-surface temperature and other paleoclimatic variables over the lifespan of a single colony. This project takes a somewhat different approach, utilizing extensive underwater drilling that penetrated through the growing reef veneer into underlying fossil reef strata and Pleistocene limestone. Recent drilling through Holocene reefs in the northeastern Caribbean allows the construction of a composite paleoclimatic record from corals whose ages span the latter half of the Holocene. Although the composite record produced in this fashion is discontinuous, it has the potential to extend the range of coral-based climatic records by orders of magnitude. The composite record also reveals coincident changes in coral reef structure accompanying climatic and sealevel changes. The resolution of the record is limited by the number, age and type of corals encountered in the drilling. Certain species of coral have known uncertainties in isotopic and elemental fractionation, and the skeletonization patterns of certain species do not lend themselves to high-resolution, sub-annual analysis.Three fossil Montastrea annularis pieces from two island-shelf study sites, Lang Bank, St. Croix, and La Parguera, Puerto Rico, were sampled using techniques that avoid mineralogic inversion and isotopic artifacts (Gill, Olson and Hubbard, 1995). Coral ages are 398(±80), 1768(±70) (La Parguera) and 5958(±90) ybp (Lang Bank) as dated by carbon-14 with tree-ring calibration. Eighteen (18) to twenty four (24) microsamples were taken per annual growth-band, with a recoverable record between one and two decades per coral. All coral samples show strong annual isotopic periodicity. Total isotopic variability ranges from -6.0 to -3.5 ‰ ¹⁸O (PDB) and from -3.1 to +2.0 ‰ ¹³C (PDB) (n=426).The isotopic chemistry of corals on Lang Bank is not likely to be affected by localized water-mass heating, runoff or other land-induced effects. Lang Bank is up-current and up-wind from the island of St. Croix, and is separated from the island by 15 km of open ocean. It is more difficult to rule out such effects at the La Parguera site. However, seasonal rainfall patterns would tend to exaggerate rather than minimize oxygen isotopic variation, and modern monitoring does not indicate meteoric or saline influences on seawater isotopic chemistry at the La Parguera site. Samples from Lang Bank and La Parguera show no statistical difference in annual average oxygen isotopic values, suggesting that average sea-surface temperatures did not change measurably at the two sites over the time range from 3960BC to AD1600. However, the 6000 ybp coral sample showed a larger range of annual isotopic excursions, implying that although average annual temperatures were similar, there was slightly greater seasonal temperature variation approximately 6000 ybp. In particular, 400 years ago the peak summer sea-surface temperatures appear to be depressed, and the minimum winter temperatures elevated, relative to the corresponding temperatures 6000 ybp. Interpreting the isotopic data strictly as a temperature signal, the data do not indicate changes in average NE Caribbean SST for the last half of the Holocene. This is surprising because the three samples span the time period thought to produce the highest temperatures (Holocene Temperature Maximum, ca. 5000-6000 ybp) and lowest temperatures (Little Ice Age, ca. 150-450 ybp) in the last 6000 years. The results tentatively support the notion that either tropical oceanic climates are insulated from climatic variability, or that the Holocene Temperature Maximum and the Little Ice Age did not affect the Caribbean. The apparent change in seasonal temperature range is potentially significant in light of the strong effect temperature maxima and minima have on scleractinian health and coral reef growth.

Gill, I, Olson, JJ & Hubbard, DK, Geology, 23, 333-336, (1995).

68