Quantifying climate-mode–driven ocean variability reveals intensified sea-level rise

Sea level is significantly modulated by various climate modes. Yet, the impacts of individual modes, their evolution, and links to present-day sea-level rise remain elusive. Here, we develop a data-driven, deep-learning-aided framework combining satellite altimetry and tide-gauge data to reconstruct seven decades (1952–2022) of global sea level. An Empirical Orthogonal Function decomposition of our reconstruction separates global sea-level rise from variability driven by the modulated annual cycle, El Niño–Southern Oscillation, Interdecadal Pacific Oscillation, and North Pacific Gyre Oscillation (NPGO), while quantifying each component. We reveal intensification across all quantified variability components. The intensity (standard deviation) of global-mean variability from summing these components nearly doubles, from 0.73 mm in 1952–1982 to 1.45 mm in 1992–2022. The corresponding global-mean sea level (GMSL) trend also nearly doubles, from 1.59 ± 0.26 mm yr⁻¹ to 3.10 ± 0.28 mm yr⁻¹. In moving-window estimates, both metrics are more pronounced since the late 1960s, and GMSL trends show relatively high correlations with the NPGO-associated variability intensity series (r ≈ 0.45–0.85; scale-dependent). We conclude that climate modes’ impacts are not static; they also strengthen alongside intensified sea-level rise, potentially reshaping the spatiotemporal characteristics of sea level under an increasingly warmer Earth.