Geochemical Heterogeneity at the Millimeter Length Scale

It is well known that even within the same locality, where water chemistry is essentially the same, the use of different coral species can lead to different

SST calibrations, presumably due to biological differences (e.g., Gagan et al., 2000; Marshall and McCulloch, 2002; Watanabe et al., 2002, 2003). Even among the same coral species from the same location, SST calibrations can differ among colonies (Allison, 1996; Wellington et al., 1996; Linsley et al., 1999; Watanabe et al., 2003). Some of these discrepancies can be explained by different sampling resolution, variations in skeletal growth-rate, and small environmental heterogeneities in the reef environment. However, millimeter-scale observations of geochemical heterogeneity within different skeletal elements of the same corallite in a given colony present an additional complication. The aragonitic skeletons of hermatypic corals are comprised of thousands of individual corallites, each with a more or less well defined theca-wall, columellae, septa, and dissepiments (Fig. 4). The theca-wall and columella grow in the vertical direction, the direction for extension. Septa grow vertically and horizontally at the same time, and dissepiments define horizontal layers within each corallite, like floors in a tall building (Barnes and Lough, 1993; Barnes et al., 1995). Differences in growth-direction and growth-rate of each skeletal element could affect the reconstructed climate signals. Land et al. (1975) and Watanabe et al. (2002, 2003) examined the isotopic heterogeneities among different skeletal components

Diploastrea

Septa

Diploastrea

Columella

10 mm

Columella

10 mm m

to cc

10 5

Distance (mm)

10 5

Distance (mm)

Polyp -

Theca wall Dissepiment -Septa

Montastrea

Montastrea

Theca wall Dissepiment -Septa

Oxygen

Oxygen

Distance (mm)

Distance (mm)

Figure 4: (a) Morphology of coral skeletal structures and (b) millimetre scale heterogeneity in isotopic compositions for different skeletal elements in Diploastrea and Montastrea (Watanabe et al., 2002, 2003).

using coral species with relatively large corallites and wide intercorallite spacing. The results suggested that there is a rather large difference in the stable isotope chemistry between vertically and horizontally growing skeletal elements. Vertically growing elements, such as the theca-wall, yield a certain stable isotopic signature, but the inclusion of horizontally growing skeletal elements change this signature significantly (Fig. 4). The coral species most commonly used for paleo-climatic reconstructions is Porites, which has relatively small corallites (0.5-1 mm) for which it is difficult to sample separately each type of skeletal element using conventional sampling techniques (e.g., dentist drill). On the other hand, a smaller corallite size allows better, more representative average compositions to be obtained. Still, it seems essential to understand the cause of the variable stable isotope signatures in different skeletal components at the millimeter-length scale.

Since late 1990s, microanalytical methods such as Laser ICP-MS and Secondary Ion Mass Spectrometry (SIMS) have been applied more or less systematically for measuring trace element of coral skeletons (Allison, 1996; Hart and Cohen, 1996; Sinclair et al., 1998; Fallon et al., 1999). However, the most striking result of these analyses was the large data scatter, often with amplitudes that were impossible to interpret as the result of changes in environmental conditions.

Two dramatic examples of mm-scale geochemical variations are illustrated in Fig. 5, which show conventional ion microprobe analyses of S18O and Sr/Ca ratios in wall segments and septa of the Porites skeleton. Both data sets show oscillating variations with amplitudes that are impossible to explain by fluctuations in SST or changes in any other local environmental parameter (Meibom et al., 2003; Rollion-Bard et al., 2003). Indeed, if converted to temperature, the 818O and Sr/Ca variations displayed in Fig. 5 would correspond to changes of more than 15°C, which is three times more than the seasonal SST variations at the sites where these corals lived. Similar, but less distinctly oscillating Sr/Ca variations were reported by recent works (e.g., Allison and Finch, 2004; Cohen and Sohn, 2004). The negative isotopic values can only be explained by supposing that coral skeleton is partially deposited according to a kinetic process. The boron isotopic measurements at micrometer scale in the same spots than oxygen analyses (Rollion-Bard et al., 2003) indicate that a variation of one pH unit in good agreement with direct microelectrodes measurements (Al-Horani et al., 2003). In addition, the d18O and Sr/Ca oscillations are characterized by an approximately monthly wavelength, which could be caused by biological processes changing in response to the lunar cycle (Meibom et al, 2003). We suggest that the pH controls the 818O of the growing carbonate skeleton through the relative fractions of dissolved carbonate species and through the kinetics of their isotope equilibration with water via hydration and hy-droxylation.

(a) 6 months of growth

Figure 5: Micro-scale heterogeneity observed in coral Porites of 818O (a; Rollion-Bard et al., 2003) and of Sr/Ca (b; Meibom et al., 2003) measured by ion microprobe.

Was this article helpful?

0 0

Post a comment