Fig. 2: Pure ENSO and TWEIO teleconnections in the reference dataset during boreal November and December.

ERA5 200 hPa geopotential height anomalies (Z200) and WAF regressed against the pure (i.e., independent of ENSO variability) TWEIO precipitation index for a November, b December. c, d as in a, b but for the pure (i.e., independent of TWEIO variability) Niño 3.4 regression. Shading indicates the regression of geopotential height anomalies (units m), while vectors represent the WAF (units m2/s2) with a unit vector of 0.2 m2/s2. The black solid boxes in a indicate the western (10°S–10°N, 40°E–80°E) and eastern (10°S–10°N, 100°E–140°E) Indian Ocean domains used to compute the TWEIO index. The black solid box in c indicates the domain (5°S–5°N, 170°W–120°W) used to compute the Niño 3.4 index. The light-gray stippling indicates the grid points in which the regression is significant at the 95% level of confidence with respect to a two-tailed t-test. The units of regression are specified per standard deviations.