Global Hotspots for Sea-Level Changes: Decadal Extremes and Uncertainties From CMIP6 Models

Abstract

Knowledge is limited regarding decadal extremes and uncertainties of sea-level change (SLC) at the regional scale, which necessitates the need for better understanding of these changes to enhance future coastal preparedness. To this end, we examined sea-level extremes for 23 World reference regions based on tide gauges (TGs) observations from 1950 to 2014. We then used these observations to evaluate the fidelity of climate models and earth system models (ESMs) participating in the Coupled Model Intercomparison Project Phase 6 (CMIP6), using trend analysis, correlation coefficient and Root Mean Square Error (RMSE) as metrics. Our findings show the spatial distribution of SLC varies between −10.0 and 6.52 mm/yr, with an area-weighted global average of 1.29 ± 0.32 mm/yr. Five regions display rapid increasing rates exceeding twice the global average: E. and C. North-America, N. Central-America, W. C. Asia, and S. E. Asia. Together, these regions constitute 26% of the total analyzed area. The CMIP6 simulations, especially ESMs, have a systematic underestimation of SLC, compared with TGs. We found poor agreement between CMIP6 simulations and TGs (weak correlation and larger RMSE) in subtropical North Atlantic regions and W. Central Asia. Our findings facilitate a multifactor hazard regional analysis that includes SLC alongside temperature, precipitation, and other parameters. It also identifies directions for future model development toward better detecting extremes and narrowing uncertainties in sea level simulations.