PACIFIC ATOLLS: SEDIMENTOLOGICAL DISCOVERIES IN ENIGMATIC 'CONGLOMERATE PLATFORM' SUPPORT HOLOCENE RAPID METER-SCALE SEA-LEVEL OSCILLATION AND TSUNAMI DEPOSITION

Pacific atoll rims comprise a 'reef flat' (separating lagoon and ocean), partly covered by tabular bodies of coral rubble ~1.5m thick jointly called 'conglomerate platform' (CP; e.g. Montaggioni et al. 2021), overlapped by sandy islets. The reef flat is dry at low spring tide (tide range 0.5-2m). Exposed CP surfaces are dark (boring algae). Each CP body is bounded by an eroding scarp. Coral clasts in the scarp are mostly 6.5-2.5kyBP.

Most workers assume the CP rubble source is the reef front. Instead proving a lagoon source: 1) well-exposed CP bodies on Google Earth images (e.g. Majuro, Tarawa, Caroline, SW Fakarava) have an ocean-facing 'tulip' shape, i.e. a thin (<25m) 'stem' extends up to 600m from the lagoon edge and widens abruptly into a 'flower' 150-250m tall, never reaching the ocean; and 2) flowers protruding in front of islets on both N and S rims of Majuro and Tarawa show cross-bed foresets (previously unreported in CPs) dipping oceanward (my 2018-19 fieldwork). Stems also show likely cross-beds, expressed as nested crescents facing in- or outward on Google Earth images.

Most authors interpret CP as inter- or subtidal storm beds. A new model follows. Around 3kyBP, the reef flat was exposed for centuries ('Pelham Bay Emergence', Fairbridge 1961). About 2.7kyBP, world sea level (SL) quickly rose ~5m in ~100y ('Abrolhos Submergence'), to ~2m above modern SL, re-flooding the reef flat, removing any sand cover (wave attack) and outpacing coral growth. After a threshold water depth was reached, the first tsunami to cross the newly submerged atoll rim induced contra-rotating vortices behind the rim's tsunami-facing sector (now an 'underwater barrier' sensu Boshenyatov & Zhiltsov 2019). The vortices ejected patch-reef talus bilaterally 'sideways' out of the lagoon.

SL later fell ~5m by ~1.7kyBP ('Florida Emergence'). Foram-rich sand exposed on the reef flat and lagoon fringe became heaped into eolian dunefields (NB ~2kyBP forams in islet sand, Kayanne et al. 2011). An ensuing SL rise of ~3m by 1.0kyBP ('Rottnest Submergence') re-flooded the reef flat, shrinking the dunefields (wave attack) to form today's islets. Where sand supply is high, islets now prograde into the lagoon.

The oscillating SL model contradicts Kiritimati's 5-0kyBP 'stable' SL curve, whose clustered samples and wide error bars permit multi-century data gaps.