Development and Problems Associated with Coral Paleo Temperature Proxies

The oxygen isotopic (S18O) and strontium/calcium (Sr/Ca) ratios of massive coral skeletons have been widely used as proxies for past changes in SST of the tropical and subtropical oceans, because both the geochemical parameters are believed to depend on the temperature of the ambient seawater.

The temperature dependency of the oxygen isotopic composition in inorganically precipitated carbonates was established by Urey (1947), but it was realized early on that the 818O of biogenic carbonates reflects a combination of environmental parameters and biological processes, so-called "vital effects'' (Urey et al., 1951; Epstein et al., 1953; McConnaughey, 1989). In spite of a biologically induced isotopic disequilibrium with respect to seawater, corals in general provide the best d18O paleo-records of ambient SST variations in waters with near constant 818O, i.e., waters in hydrologic balance (e.g., Leder et al., 1996; Wellington et al., 1996). (Alternatively, corals can provide a record of changes in hydrologic balance in regions where the temperature is nearly constant; e.g., Cole and Fairbanks, 1990). However, seasonal variations in d18O of the seawater on the reef are often relatively large (Leder et al., 1996), and in some cases, the effect of seawater S18O variations on the d18O vs. SST calibration can be quantified (McCulloch et al., 1994; Juillet-Leclerc and Schmidt, 2001). Modern approaches to the construction of 818O paleo-environmental records typically involve coring of living coral heads and calibration to (5-10 year) instrumental temperature records obtained in close vicinity to the coral (Stephans et al., 2004).

Temperature dependence of the Sr/Ca ratio in aragonite was demonstrated by Smith et al. (1979). However, initially the analytical errors associated with the analyses of coral Sr/Ca ratios were large (72°C, if converted to temperature), limiting the use of the coral Sr/Ca temperature-proxy to localities where temperature variations are significantly larger. Later, Beck et al. (1992) developed isotope dilution and thermal ionization mass spectrometry (TIMS) techniques that allowed the measurement of Sr/Ca ratio with an analytical error corresponding to less than 70.5°C for reconstructed SST. In contrast to the isotopic composition of seawater, the Sr/Ca ratio of seawater has been assumed to be essentially constant because of the long residence times of strontium in the ocean (5.1 x 106 yr; Broecker and Peng, 1982). However, more recent observations of the ocean Sr/Ca ratio from different depths and localities raise some doubt about the general validity of this assumption (de Villiers et al., 1994; de Villiers, 1999).

McCulloch et al. (1994) assessed d18O fluctuations in seawater by removing the temperature dependency from the 818O signal using the calibrated Sr/Ca vs. SST relationship. This method was extended to reconstructions of past sea surface salinity (SSS) (Gagan et al., 1998; Ren et al., 2002) assuming a linear relationship between the S18O of seawater, SSS, and SST. With the development of automated methods for rapid sampling and analyses of carbonate powders, different methods for reconstruction of SST have been applied to a century-long coral records covering most of the tropics (Fig. 1), as well as to climatically important periods in the more distant past, such as the last glacial maximum (e.g., Guilderson et al., 1994), Little Ice Age (Watanabe et al., 2001), and last interglacial warm period (Tudhope et al., 2001). At the same time, it has become clear that coral-based paleo-climatic reconstructions suffer from problems that are not only related to the analytical challenge of obtaining high-quality 818O and Sr/Ca data. Fig. 2 shows the large discrepancy among reported modern calibrations both for coral d18O (Fig. 2a) and Sr/Ca (Fig. 2b). These discrepancies suggest that simple application of different coral species, localities, and methods can lead to large discrepancies among different reconstructions of past SST (Marshall and McCulloch, 2002; Watanabe et al., 2002).

http://www.ngdc.noaa.gov/paleo/corals. html)."/>
Figure 1: Coral records of 818O and Sr/Ca ratio from modern corals. Thick lines represent 5 years running averages. Data are available at the World Data Center for Paleoclimatology (http://www.ngdc.noaa.gov/paleo/corals. html).

os oo

O oo

Synthetic Aragonite ■ Tarutani et al. (1969)

Synthetic Calcite O'Neil et al. (1969).

Synthetic Aragonite ■ Tarutani et al. (1969)

Synthetic Calcite O'Neil et al. (1969).

O oo

2 Mitsuguchi et al. (1996)

Temperature (°C)

Corals

Temperature (°C)

2 Mitsuguchi et al. (1996)

3 de Villiers et al. (1994)

20 25 30

Temperature (°C)

Figure 2: Diagrams showing the disagreement between different calibrations of SST vs. (a) the oxygen isotopic composition in Porites lutea and (b) the Sr/Ca ratio in coral Porites. Each calibration has been calculated from samples collected along main growth axis, by comparing the geochemical measurement with the SST variability.

3 de Villiers et al. (1994)

20 25 30

Temperature (°C)

Figure 2: Diagrams showing the disagreement between different calibrations of SST vs. (a) the oxygen isotopic composition in Porites lutea and (b) the Sr/Ca ratio in coral Porites. Each calibration has been calculated from samples collected along main growth axis, by comparing the geochemical measurement with the SST variability.

Was this article helpful?

0 0

Post a comment