High Precision Thorium-230 Ages of Corals and the Timing of Sea Level Fluctuations in the Late Quaternary

Citation

Edwards, Richard Lawrence (1988) High Precision Thorium-230 Ages of Corals and the Timing of Sea Level Fluctuations in the Late Quaternary. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/S2FW-0463. https://resolver.caltech.edu/CaltechETD:etd-10112005-103528

Abstract

Mass spectrometric techniques for the measurement of ²³⁰Th and ²³⁴U have been developed. These techniques have made it possible to reduce the analytical errors in ²³⁰Th dating of corals using very small samples. Samples of 8 x 10⁷ atoms of ²³⁰Th can be measured to an accuracy of ±130% (2 sigma), 6 x 10⁸ atoms of ²³⁰Th can be measured to an accuracy of ±29%, and 3 x 10¹⁰ atoms of ²³⁰Th can be measured to an accuracy of ±2%. The time range over which useful data on corals can now be obtained ranges from 15 to 500,000 years. The error in age (based on analytical error) for a sample that is 18 years old ±3 years (2 sigma). The error is ±5 years at 180 years, ±44 years at 8294 years, and ±1 ky at 123.1 ky. For young corals, this approach may be preferable to ¹⁴C dating.

Fluctuations in climate result in changes in sea level because the ice stored in continental glaciers is ultimately derived from the ocean. Certain species of coral grow close to the sea surface. Fossils of these species therefore record the former height of the sea surface. The precision with which the age of a coral can now be determined makes it possible to determine, with some precision, the timing of sea level fluctuations in the late Quaternary. This record will allow a critical test of the Milankovitch hypothesis, which predicts the timing of Pleistocene climate fluctuations from changes in the distribution of solar insolation that result from changes in the earth's orbital geometry. Analyses of a number of corals that grew during the last interglacial period yield ages of 122 to 130 ky. The ages coincide with or slightly postdate the summer solar insolation high at 65°N latitude, which occurred 128 ky ago. This supports the idea that changes in Pleistocene climate can be the result of orbital forcing.

Apparent fluctuations in sea level recorded on tectonically active shorelines are the result of both sea level change and vertical tectonic movement. If the record of sea level change is known (e.g., from the coral record in a stable area), this record can be subtracted from the record of apparent sea level change, in the tectonically active area, to yield a record of vertical tectonic movement. The precision with which coral ages can now be determined may allow us to resolve the ages of individual coseisimic uplift events and thereby date prehistoric earthquakes.

This possibility has been examined at two localities, northwest Santo Island and north Malekula Island, Vanuatu. Previous work (Taylor et al., 1980, 1985a, 1987) showed (using the counting of annual growth bands to determine ages) that the tops of partially emerged coral heads at each locality died at the same time as the last major earthquake at each locality (M_S = 7.5, 1973, on northwest Santo; and M_S = 7.5, 1965, on north Malekula). It was concluded that the tops of these coral heads were killed by coseismic uplift. At each locality, there were also completely emerged coral heads, which were inferred to have been killed by earlier coseismic uplift events. These could not be dated by growth band counting because the coral heads were completely dead.

The accuracy of ²³⁰Th ages of very young corals was tested by dating portions of three corals whose ages were known from the counting of growth bands. Within analytical error, the ²³⁰Th ages were the same as the growth band ages for all three samples (dates of growth by counting growth bands - A.D. 1971 to 1973, A.D. 1935 to 1939, and A.D. 1804 to 1810; dates of growth from ²³⁰Th measurements - A.D. 1969 ± 3, A.D. 1932 ± 5, and A.D. 1806 5 [2 sigma ±1) demonstrating that the ²³⁰Th ages were accurate.

The ²³⁰Th growth dates of the surfaces of adjacent emerged coral heads, collected from the same elevation (1.2 m) on northwest Santo Island, were, within analytical error, identical (A.D. 1866 ± 4 and A.D. 1864 ± 4). This indicates that the corals died at the same time and is consistent with the idea that they were killed by coseismic uplift. Similar adjacent coral heads on north Malekula Island yielded ²³⁰Th growth dates of A.D. 1729 ± 3 and A.D. 1718 ± 5. The ages are similar but analytically distinguishable. The difference may be due to erosion of the outer, younger, portion of the latter coral head. Using the date of the large historical earthquake at each locality and the ²³⁰Th growth date of the emerged corals at each locality, recurrence intervals of 108 years for northwest Santo and 236 years for north Malekula are calculated.

This experiment has shown that it is possible to date corals that grew in the past several centuries to accuracies of ±3 to ±5 years (2 sigma). The main problems with applying this approach to determine seismic histories will be associated with the preservation of fossil corals that have been killed by coseimic uplift and the ability to identify such features in the field.

Thesis Files