Doctor of Philosophy
School of Geosciences
Smithers, Scott Geoffrey, The elevation and surface morphology of microatolls on the Cocos (Keeling) Islands, Indian Ocean, Doctor of Philosophy thesis, School of Geosciences, University of Wollongong, 1997. http://ro.uow.edu.au/theses/1988
This thesis examines the elevation and surface morphologies developed by massive Porites microatolls growing in reef flat, interisland passage and lagoonal habitats around the Cocos (Keeling) Islands, Indian Ocean. It has long been known that corals adopt the microatoll form near to the sea surface, and microatolls are widely recognised as important and relatively precise indicators of sea-level position, particularly through the mid to late Holocene. Recently it has also been suggested that the upper surface microtopographies of modern microatolls may yield valuable information about recent sea-level change. However, rarely has the elevation of microatolls been accurately surveyed to a known datum, or the microtopographies of their upper surfaces accurately measured. This thesis has addressed this shortcoming by accurately surveying the upper limit to coral growth around the rims of 282 microatolls from 19 separate sites around the Cocos (Keeling) Islands, Indian Ocean to a known datum (MSL at the H o m e Island tide-gauge), and by systematically sampling and documenting the morphological development of more than 120 microatolls. Relationships between microatoll surface form and sea-level and environmental change were also explored.
The height of living coral (the 'aHLC') around individual microatolls was found to vary 'intrinsically' by approximately 2cm, irrespective of where the microatoll grew, representing the precision level with which the a H L C and the constraining water level (ra) are typically coupled. The aHLC did not vary systematically around microatoll rims; the majority of microatolls on Cocos have relatively horizontal upper surfaces. Separate elevational groups were statistically discriminated at most survey sites, and were attributed to the combined occurrence of microatolls that were confined at sea-level and/or one or more ponded water levels, or because some microatolls were not water-level limited. Reef flat microatolls were typically constrained at greatest depth, with interisland passage microatolls gradually becoming higher as the passages shallowed toward the lagoon. The aHLC of lagoonal microatolls was variable, and appeared to be controlled by the depth of the underlying substrate. The upper limit to coral growth around microatolls freely connected to the open sea at low tide was restricted at a level approximately midway between M H L W and MLLW, though their actual elevations varied around the atoll, reflecting variations in reef morphology and tidal hydrodynamics.
Most microatolls had statistically similar microtopographies across separate growth axes. Lagoonal and reef flat microatolls were the most consistently symmetric, and typically developed prominent concentric 'bumps' over their upper surfaces. In contrast, interisland passage microatolls were least often symmetric, and typically developed flat upper surfaces with little topographic relief, ft was concluded that surface symmetry is promoted in freedraining reef flat and lagoonal habitats where the potential amplitude of water-level change is large relative to both the noisiness of the water level surface and the range of intrinsic aHLC variability, because a strong and persistent water-level change signal is produced that is reflected in the direction, if not the magnitude of rim growth. Where the microtopographies of individual microatolls are symmetric, adjacent microatolls were also most likely to be similar. The similarity of microatoll microtopographies in like habitats around the atoll was generally low. It was speculated that this divergence reflects the variety of physiographic conditions possible within each habitat type, and the complexity of biological responses to this diversity.
The surface microtopographies of two long-lived open water reef flat microatolls from different sites, dominated by broad undulations of 5-10cm amplitude developed over 18-20 year cycles, were determined to be similar statistically. The independent development of this congruent morphology by these separated microatolls implies the influence of a common environmental signal. Statistical correlations between microatoll surface morphology and the short and fragmented tide-gauge record available for Cocos are not strong, but this was expected given the extent and quality of the tide-gauge data and the fact that microatoll response to water-level change is biologically mediated and unlikely to be sensitive to shortterm or subtle fluctuations. Nevertheless, a visual comparison between microatoll surface microtopography and the tide-gauge record shows similar trends. Microatoll microtopography and a range of other environmental variables for which instrumental data were available were very weakly correlated. An intriguing temporal coincidence between the 18.6 year cycle of lunar tides (related to slight variations in the tilt of the moon's orbit relative to the sun's equator) and the broad undulations that dominate the microtopographies of large open water reef flat microatolls on Cocos was identified, suggesting the possibility that the broad undulations develop in response to tidal adjustments to the 18.6 year lunar cycle. A tentative link between ENSO events and low water levels on Cocos may also be speculated from the timing of stronger events and marked falls in the upper limit to coral growth represented by depressions over the microatoll plane, though this hypothesis requires further investigation. It was concluded that the surface morphologies of open water reef flat microatolls on Cocos were principally influenced by sea-level, though their microtopographies are more likely to preserve filtered, longer-term sea-level trends than interannual variation. The surface microtopographies of large microatolls on Cocos are clearly dominated by the cyclic undulations described above, suggesting that the water level that constrains upward coral growth on Cocos has fluctuated through a series of regular cycles over at least the last century. The surface form of microatolls on Cocos does not support the 10-20cm rise in global sea level over the last one hundred years, indicated in aggregated tide-gauge records, but instead indicates sea level at Cocos has undergone little net change over this period. It was concluded that microatoll upper surfaces preserve subdued, low temporal resolution records of sea-level change, and represent a valuable source of retrospective sea-level data in areas where tidegauge records are short or lacking.