Revisiting tectonic corrections applied to Pleistocene sea-level highstands

Abstract

Tectonic displacement contaminates estimates of peak eustatic sea level (and, equivalently, minimum continental ice volumes) determined from the elevation of Quaternary interglacial highstand markers. For sites at which a stratigraphic or geomorphic marker of peak Marine Isotope Stage (MIS) 5e sea level exists, the standard approach for estimating local tectonic uplift (or subsidence) rates takes the difference between the elevation of the local highstand marker and a reference MIS 5e eustatic value, commonly chosen as +6 m, and divides by the age of the marker. The resulting rate is then applied to correct the elevation of all other local observed sea-level markers for tectonic displacement, including peak highstands of different ages (e.g., MIS 5a, MIS 5c and MIS 11), under the assumption that the tectonic rate remained constant over those periods. This approach introduces two potentially significant errors. First, the peak eustatic value adopted for MIS 5e in most previous studies (i.e., +6 m) is likely incorrect. Second, local peak sea level during MIS 5e is characterized by significant departures from eustasy due to glacial isostatic adjustment in response to both successive glacial–interglacial cycles and excess polar ice-sheet melt relative to present day values. We use numerical models of glacial isostatic adjustment that incorporate both of these effects to quantify the plausible range of the combined error and show that, even at sites far from melting ice sheets, local peak sea level during MIS 5e may depart from eustasy by 2–4 m, or more. We also demonstrate that the associated error in the estimated tectonic rates can significantly alter previous estimates of peak eustatic sea level during Quaternary highstands, notably those associated with earlier interglacials (e.g., MIS 11).