Imprint of relative sea-level histories on last interglacial coral preservation

Summary Fossil corals are commonly used to reconstruct Last Interglacial (∼125 ka, LIG) sea level. Sea-level reconstructions assume the water depth at which the coral lived, called the ‘relative water depth’. However, relative water depth varies in time and space due to coral reef growth in response to relative sea-level (RSL) changes. RSL changes can also erode coral reefs, exposing older reef surfaces with different relative water depths. We use a simplified numerical model of coral evolution to investigate how sea-level history systematically influences the preservation of corals in the Bahamas and western Australia, regions which house > 100 LIG coral fossils. We construct global ice histories spanning the uncertainty of LIG global mean sea level (GMSL) and predict RSL with a glacial isostatic adjustment model. We then simulate coral evolution since 132 ka. We show that preserved elevations and relative water depths of modeled LIG corals are sensitive to the magnitude, timing, and number of GMSL highstand(s). In our simulations, the influence of coral growth and erosion (i.e. the ‘growth effect’) can have an impact on RSL reconstructions that is comparable to glacial isostatic adjustment. Thus, without explicitly accounting for the growth effect, additional uncertainty is introduced into sea-level reconstructions. Our results suggest the growth effect is most pronounced in western Australia due to Holocene erosion, but also plays a role in the Bahamas, where LIG RSL rose rapidly due to the collapsing peripheral bulge associated with Laurentide Ice Sheet retreat. Despite the coral model's simplicity, our study highlights the utility of process-based RSL reconstructions.