The Effect of Internal Climate Variability on Spatial and Temporal Patterns of Sea-Level Rise Variations and Projections

Abstract:

Natural (or internal) climate variability can influence spatial patterns of sea-level variations on interannual-to-interdecadal time scales, through redistribution of ocean heat and dynamical changes in ocean circulations and surface winds. These variations can impact analyses of past and potential future sea-level rise, particularly when considering upper bound projections on regional scales. Here we analyze spatial and temporal sea-level variations due to steric and dynamic effects, using results from a 50-member climate change ensemble experiment that samples internal variability of the coupled ocean-atmosphere system. We use the low-resolution configuration of the fully-coupled Community Earth System Model (CESM), comprised of transient hindcasts and projections (1850-2100) using historic and RCP8.5 forcings. The transient simulations are initialized from unique model states, sampled from a ~10,000 year fully coupled unforced equilibrium simulation. This approach enables us to examine the effect of initial conditions uncertainty (or internal variability) within the coupled system, including effects of the full ocean which has a longer dynamical time scale than the atmosphere. We find that the internal variability within the coupled ocean-atmosphere climate model significantly affects spatial patterns of dynamic and steric sea-level changes across multiple timescales. We evaluate model skill using the historical record of past trends and variations across multiple scales, and we explore the implications of these results for sea-level rise projections. A key result is that natural variations of sea-level change can significantly impact upper-bound projections at regional scales, which can potentially impact risk assessments and adaptation strategies. More generally, these results point to the potential importance of the role of ocean dynamics and memory in influencing sea-level rise variations across multiple spatial and temporal scales.