Positions of active regions 1, 2, 4, 5, and 6 deduced from the seismic maps, with respect to their actual positions deduced from the SO/PHI magnetograms (a). The active region coordinates are computed using a center-of-gravity method, applied to each observable (Φ or B_los) within the region areas. For the SO/PHI magnetograms, the active region areas correspond to the red contours in Figs. 2 and A.1 (20 G contours from the smoothed magnetograms). For the helioseismic data, a threshold of 2σ_Φ is used to determine the active region areas. The blue ellipse shows the probability density function of the data points at half maximum, assuming a bivariate normal distribution. The dashed circle with a diameter 4° corresponds to the resolution limit of the seismic observations. Active region areas inferred from helioseismology versus areas measured from the SO/PHI magnetograms (b). The linear fit gives a slope of 0.98 (solid blue line). The dashed line is the 1:1 diagonal.