The mid-Holocene sea-level highstand refers to higher-than-present relative sea levels (RSLs) in far-field regions between 7000 and 4000 years ago because of equatorial ocean syphoning and continental levering. But the timing, magnitude, and spatial variability of the highstand are uncertain and the highstand parameterization in Glacial Isostatic Adjustment (GIA) modelling is understudied. Here, we use the RSL records of Southeast Asia to investigate the sensitivity of the mid-Holocene highstand properties to Earth and ice model parameters, including lithospheric thickness, mantle viscosity (both 1D and 3D), and deglaciation history of Antarctica and global ice sheets. We found that the Earth model variation only affects the magnitude of the mid-Holocene highstand unless extraordinary low upper mantle viscosity is used. The timing of the highstand moves towards present and there is an absence of the highstand if upper mantle viscosity is < 4.0 × 1019 Pa s or ≤ 1.0 × 1019 Pa s, respectively. Ice model variation changes both the timing and magnitude of the mid-Holocene highstand. Delaying the ice-equivalent sea level will shift the timing of the highstand later and result in a lower highstand magnitude. We produced a mid-Holocene highstand “treasure map” that considers topography change and accommodation space to guide future RSL data collection efforts in Southeast Asia. The highstand “treasure map” indicates that the northeast and central west coast of Malay-Thai Peninsula, east coast of Sumatra, north coast of Java, and southwest coast of Borneo are very likely (90% probability) to preserve mid-Holocene highstand evidence.
Read full abstract