9710005 Fletcher The submerged slopes of the northern main Hawaiian islands of Kauai and Oahu are marked with a flight of terraces that are fossil carbonate reef tracts dating from sea-level stands during the late Quaternary. The seaward-facing front of each reef terrace is a near vertical wall that displays drowned shoreline features such as intertidal notches, beachrock, and growths of fossil coral (Fletcher and Sherman, 1995). Of special note are the submerged intertidal notches that are found island-wide at predictable depths with well-preserved overhanging visors. These, and related paleoshoreline features, indicate the occurrence of abrupt jumps in relative sea-level, which, in the absence of a local mechanism for coseismic displacement, are best explained as the product of glacial melt-water pulses. The occurrence of these shorelines is reminiscent of melt-water pulses identified in cores of fossil coral from Barbados (Fairbanks, 1989) and elsewhere (Edwards et al., 1993; Bard et al., 1996; Montaggioni, submitted). Other drowned shorelines have also been described, such as those found in south Florida (Locker et al., 1996) and New Zealand (Carter et al., 1986). Four drowned shorelines are identified in Hawaii. The deepest (approx. -110 to -120 m) probably correlates to the glacial lowstand of sea level which occurred around the same time as Heinrich Event H-2 (?). Drowned shorelines at depth ranges of -90 to -95 m and -58 to -60 m are approximately in the same depth range as Barbados melt-water pulses IA and IB, resp., however there are discrepancies of several meters in cases. We speculate that the shoreline at -58 to -60 m is cut into a reef dating from the last major interstadial (Stage 3). The shallowest shoreline, -24 m, correlates to records of a prominent melt-water pulse in the North Atlantic (Keigwin and Jones, 1995) and a drowned shoreline on the Great Barrier Reef Shelf of Australia (Larcombe et al., 1995). This shallowest shoreline is carved into fossil carbon ate of ca. 190 to 240 kyrs age (Stage 7). Additionally, the three shallowest shorelines correlate in their depth ranges with the controversial catastrophic rise events (CRE's) postulated by Blanchon and Shaw (1995) in their analysis of the Barbados coral sequence. Hence, there is a rationale for interpreting the Hawaiian submerged shoreline series as an archive of abrupt sea-level events during the post-glacial and early Holocene interval. However, the possibility of a correlation (if any) to other global records is untested and requires a detailed examination of various types of cored, datable samples to determine the age of drowning. Major uncertainty exists regarding the true nature of the relationship (if any) between these features and the stratigraphic record of sea-level movements from the tectonically active margins of Barbados and Papua New Guinea, the stable shelf of Tahiti, or the marine stratigraphy from the North Atlantic that identifies episodes of ice-rafted debris (Bond and Lotti, 1995). Further, because the Hawaiian record offers the possibility of better constraining the exact position of paleointertidal positions through analysis of the drowned notches (both msl and mllw correlate to aspects of notch morphology), it will provide an improved constraint on the behavior of sea level during two specific times of the post-glacial and Holocene interval. Hawaii is an important station in establishing whether the climatic record of the North Atlantic is of regional importance or carries a global signature, and if global, its influence on eustatic sea level. The role of sea level in melt-water events of the last deglaciation has not been well-constrained with regard to magnitude, chronology, or impact on the coastal/shelf system. MacAyeal (1993) calculates that a purge phase of Heinrich event oscillations of the Laurentide ice sheet is capable of producing a 3.5 m shift in the position of global sea level within 250 yrs. However, Clark et al(1996) propose the Antarctic ice s heet as the source for melt-water IA. Hence, the position of abrupt sea-level events in the chain of process and response is still very much ill-defined. The model of Blanchon and Shaw (1995), although derived from an interpretation of geological records, still requires additional testing and extension to the Pacific Basin. Important questions have been raised by the new records from Tahiti regarding the occurrence of meltwater events and climatic episodes (Bard et al., 1996). Questions also remain regarding the relative timing of Heinrich Events, melt-water pulses, and sea-level movements (Clark et al., 1996) that require additional dating before cause and effect relationships, and sources, are fully understood. An important missing component of deep-sea records of ice-rafting events and fresh-water discharge is the role of sea level...a component that cannot be constrained by those records. Dating an erosional notch is not easy. Research will focus on the two shallowest drowned shorelines, -24 m and -58 to -60 m. We will use datable carbonate records to define sea-level chronology and position. The age of cored coral reef sequences that are post-glacial and Holocene in age will constrain the back-stepping history of accretion. Fossil intertidal faunal assemblages preserved on the face of drowned notches, and perhaps contemporaneous beachrock exposures, will all be hydraulically cored by divers to obtain samples for age-dating as proxies for past sea level. Coral grew beneath the seas that cut the drowned shorelines, and later it grew above. Exposures of that coral are prevalent at the base and upon the visors of the paleoshorelines. Aragonitic samples (determined with XRD) from cores of these exposures, early Holocene and late Pleistocene in age, will be dated at the SOEST-TIMS facility (run by K.Ruben, UH faculty member) and at the University of Arizona accelerator radiocarbon facility (G.S. Burr has agreed to participate as geochronologist). Samples from within the arc of the paleo-notches will also be petrographically analyzed for cementation histories indicating vadose-zone diagenesis (L. Montaggioni has agreed to participate). We will also sample for stable light isotopic content (B. Popp, UH faculty member runs the UH stable light isotope mass spec. facility and has agreed to participate). Likewise, cores of coral successions landward of the drowned shoreline, marking the final position of sea level following its abrupt movement, will be chronologically, petrographically, and geochemically analyzed and described. The goal for most cores will be to penetrate the entire Holocene and post-glacial sequence present at each drill site. Hence, we will employ the concept of "basal dates" where the post-glacial/pre-glacial contact is marked by a subaerial surface characterized by diagenetic alteration. Basal dates marking the first incidence of newly risen (and stabilized) sea level are an important and recognized sea level index point that has served the tidal marsh stratigraphy community of the passive margin coasts of the U.S. in their investigations of relative sea-level history in the middle and late Holocene. In our case, we will determine the distribution of basal coral dates as they relate to sea-level movements responsible for the drowned notches. As a consequence of our drilling the Holocene section, we will also obtain samples of the fossil reef tracts that form the stair-step bathymetry of Oahu. Dates (TIMS-HAS) of pristine aragonite in these Hawaiian reefs will provide improved understanding of sea-level movements at the end of the last glacial and during the early phases of the present interglacial. We have developed the capability of coring at these depths during the course of our project EAR-9317328 using oxygen-rich breathing gases for decompression and carefully controlled safety procedures employing redundant breathing
