Fig. 3: Regression coefficient maps from the multiple variable linear regression (MVLR) analysis. | Nature Communications

Fig. 3: Regression coefficient maps from the multiple variable linear regression (MVLR) analysis.

From: Reconciling global mean and regional sea level change in projections and observations

Fig. 3: Regression coefficient maps from the multiple variable linear regression (MVLR) analysis.

Colored circles indicate significant regression coefficients (90% confidence level) of tide-gauge data (1970–2018) related to a the El Niño—Southern Oscillation (ENSO), b the Pacific Decadal Oscillation (PDO), c the North Atlantic Oscillation (NAO) (all with the same colour scale), d the Indian Ocean Dipole (IOD), e linear trend, f the Southern Annular Mode (SAM), and g acceleration (twice the quadratic term). Regression coefficients of tide-gauge records, which are not significant at 90% confidence level are denoted as cross symbols (hereafter). The corresponding coefficient contours are based on the Ocean Reanalysis System 5 (ORAS5) ocean-reanalysis over the same period. Stippling indicates where the coefficients are statistically significant at the 90% confidence level for ORAS5 reanalysis.

Back to article page