Data and analysis report: Manihiki and Rakahanga, Northern Cook Islands - for February and October/November 2004 research trips

摘要

The lagoon flushing mechanisms were investigated for the South Pacific lagoon atolls of Manihiki and Rakahanga, located in the Cook Islandsu27 northern group. This investigation included two research trips to the larger atoll of Manihiki and, during the first visit there, Rakahanga was included in the field measurements. The key field measurements included for identifying the lagoon flushing process were: -centimetre accuracy water surface levels for -- ocean -- lagoon -- inflow and outflow reef tops - decimetre per second accuracy currents for inflow and outflow reef tops The investigation revealed that the lagoon flushing is driven by ocean waves pumping water into the lagoon via those reef tops which face the larger waves. At the same time water will usually flow out on the other side of the atoll. The water level in the lagoon is above the ocean at any particular time and the magnitude of the overheight correlates closely with the height of the waves on the exposed side. Title page figure A illustrates this process with the resulting one-way flow transporting water volume through the lagoon. The quantitative field measurements are supported by visual observations of the water level on the inflow reef seen to be persistently above the lagoon while, at the outflow reef tops, the water passes through critical depth at the atoll edge. Additionally, two simple mathematical models developed during the investigation and tested against the field measurements support wave-driven flushing, modulated by the ocean tide. Finally, with no true deep reef pass, the wave pumping process explains the rapid rise in mean lagoon level that occurs as the wave conditions strengthen (second research trip measurements). This flushing mechanism in operation at both Manihiki and Rakahanga is qualitativly different to tidal flushing, which potentially can increase the mean lagoon water surface levels above the mean tidal level through non-linear feeding channel conveyances (from bed friction and cross sectional area variations) and rapid changes in storage surface area with water level (i.e. tidal flats in estuaries). However, these two processes are unable to lift the instantaneous lagoon water surface level above the ocean tide level, as measured at Manihiki and Rakahanga. Title page figure B shows qualitatively, the tidal flushing scenario if these lagoons were connected to the ocean via multiple deep and wide channels. The lagoon dynamics typical of tidal flushing are oscillating flow patterns; lower mean water surface levels; large water surface level variations; and water surface level variations are contained within the ocean tide variations. Using field measurements and several simplifying assumptions, the lagoon turn over time during average wave energy flux conditions is 50 days. The very best lagoon turn over time from tidal flushing using several very large reef passes is 62 days. With wave pumping being significantly more efficient at flushing the lagoon, flushing could be improved by increasing the reef top length. At Manihiki, this could be achieved by removing parts of the ancient reef situated along the southern coast where the reef crest currently is about 1.5 m above mean sea level. The present reef edge should remain undisturbed so as to maintain the current present lagoon water level [see Callaghan et al., 2006 for details]. The wave pump and tidal flushing processes are not mixable, as the connecting channels would short circuit the wave pumped inflow directly out to the ocean, similar to the coastal rip cell scenario. Hence, the persistent one-way flow pattern would not be established and flushing would be limited to near reef tops.