The accuracy of electromagnetic-inductive (EM) sea-ice thickness measurements performed from a moving icebreaker is evaluated. Ice thickness distributions and ridge statistics are derived. In the Bellingshausen Sea the ice was moderately deformed with mean and modal ice thicknesses of 1.3 and 0.9 m, respectively. In contrast, the heavily deformed ice in the Amundsen Sea had mean thicknesses between 2.3 and 3.1 m, with modes between 1.6 and 2.7 m. By means of the measured apparent electrical conductivity on ice of known thickness, the conductivity of the perennial summer ice was found to be significant and could not be neglected in the calibration of EM measurements. The high conductivity is mainly due to the widespread occurrence of seawater-filled gaps close to the ice surface. The system, consisting of commercially available instruments, yielded reliable mean thickness estimates. This is shown by comparison with drill-hole measurements and visual observations. However, the thickness profiles appear highly low-pass filtered compared to their surface height profile determined with a laser altimeter. Thus the system possesses a limited lateral resolution such that the thickness of single pressure ridges cannot be resolved. This is only in part attributed to the trade-off between increasing instrument height and decreasing lateral resolution of EM measurements. The low resolution is mainly due to the particular EM instrument, as its sensitivity is highly reduced if used far above the water surface, and it has a comparatively long time constant of 1 s. Additionally, the water-filled gap in the ice generally reduces the sensitivity of the measured apparent conductivity to ice thickness changes.