Extended Data Fig. 5: Robustness of ice sheet projections under NDCs to ice sheet/climate model simulation selection and treatment.
From: Projected land ice contributions to twenty-first-century sea level rise
a, Greenland. b, West Antarctica. c, East Antarctica. d, Antarctic Peninsula. Box and whiskers show [5, 25, 50, 75, 95]th percentiles. Indices refer to test (see Extended Data Table 4). Robustness test 1, default; 2, higher-resolution ice sheet models; 3, ice sheet models with the most complete sampling of uncertainties (10 models for Greenland, four for Antarctica); 4, single ice sheet model with the most complete sampling of uncertainties and (coincidentally) high sensitivity to retreat or basal melting parameter. Antarctic regions only: robustness test 5, alternative single ice sheet model with nearly as complete sampling but low sensitivity to basal melt parameter; 6, ice sheet models with the highest sensitivity to basal melt parameter; 7, climate models that lead to highest sea level contributions. 8, ice sheet models with 2015–2020 mass change in the range 0–0.6 cm SLE; 9, only ice sheet models that use the standard ISMIP melt parameterizations; 10, higher basal melt value assigned to ice sheet models that do not use the standard ISMIP6 melt parameterizations.
