High‐Resolution Dynamical Downscaling Refines the Time of Emergence of Sea Level Rise in the Northwestern Pacific

Abstract

Sea‐level rise is a major consequence of anthropogenic climate change, yet the timing when regional sea‐level change becomes detectable above natural variability depends on internal ocean dynamics. Most estimates of the time of emergence (ToE) rely on coarse‐resolution global climate models (GCMs), which may underestimate mesoscale and boundary‐current variability in dynamically active marginal seas. In this study, we estimate the ToE of sea‐level rise in the Northwestern Pacific marginal seas using high‐resolution (1/20°) dynamical downscaling with the Regional Ocean Modeling System for four Coupled Model Intercomparison Project Phase 6 (CMIP6) GCMs under Shared Socioeconomic Pathways through 2100. Evaluation of historical simulations indicates that the regional simulations better reproduce observed sea surface height variability associated with strong boundary currents and mesoscale activity than coarse‐resolution GCMs. Although forced sea‐level rise is broadly comparable between modeling frameworks, emergence occurs later in the regional simulations because stronger internal variability reduces the signal‐to‐noise ratio. Delays are largest in the deep East Sea/Sea of Japan (EJS), where steric variability dominates total variability. The cumulative emergence analysis shows that 75% of grid cells exhibit emergence ∼6–9 years later in the EJS, whereas differences over the shallow Yellow and East China Seas are minimal (−1 to +1 year). These results indicate that resolving mesoscale and boundary‐current variability strongly influences the detectability of climate‐driven sea‐level rise in dynamically active marginal seas, whereas its influence is limited over shallow shelf seas. Consequently, high‐resolution regional modeling improves sea‐level projections and coastal risk assessment in dynamically active marginal seas.