Corals have the potential to record several centuries of highly detailed environmental information in the form of chemical proxies such as trace elements and stable isotopes. Massive corals commonly lay down density bands that are used as annual markers, allowing time-series records to be produced. In addition, Montastraea faveolata, a common Caribbean coral species used for proxy records, apparently erects its skeleton in two phases, with a second phase of skeletonization producing micro-scale banding that may reflect daily cycles of carbonate secretion. While this raises the possibility of producing proxy records of daily environmental variation, it also introduces uncertainty as we seek to sample at progressively finer scales until we better understand the mechanisms and timing of skeletonization.
A scanning electron microscope (SEM) and backscattered electron imaging (BEI) study of Montastraea faveolata reveals that it produces a complex pattern of microbands three orders of magnitude smaller than the density bands commonly used for proxy records. Etched, polished sections of coral skeleton show that the microbands consist of shingled aragonite crystallites oriented at high angles to the main axis of colony growth. The microbands are 1 to 10 μm in width and suggest a second phase of skeletonization that occurs later than the main axis of corallite extension. Individual imaged crystallites show a stepped increase in crystal breadth that could either be real or an artifact of sample preparation related to differential solubility.
BEI analysis reveals that the bands of crystallites contain cyclical alternations of elemental abundance. High-resolution electron microprobe traverses (1-μm beam diameter, 1-μm step size) reveal that the elemental alternations are controlled at least in part by variations in Sr/Ca within individual crystallites. Sr concentration in the microbands ranges cyclically between 9,800 and 10,800 ppm.
The microbands appear to fill in the pore space between thecae (corallite walls) and dissepiments (corallite "floors"). This secondary skeletonization process appears to be responsible for the production of the larger-scale annual high-density bands via destruction of pore space. The possibility exists that the crystallite microbanding and the chemical alternation within it is driven by external cyclicity such as diurnal temperature variation. If this is the case, then the direction and timing of this microscale skeletonization could interfere with the collection of proxy data within the high-density band. The secondary skeletonization would likely cause "blurring" of the chemical signal and be responsible for chemical noise in proxy patterns produced by analytical techniques such as ion microprobe analysis. As the resolution of analytical techniques improves, these problems would likely be exacerbated.
On the other hand, the possibility exists for extremely high-resolution sampling within the high-density band to reveal daily environmental variation and short-lived pulse events presently difficult to detect in coral records. It is also possible that the chemistry and fabric of the microbanding patterns could be used as an indicator of the degree of diagenetic alteration, an important consideration as fossil corals are used increasingly to study paleoclimatic conditions.