Although it is known that satellite data are useful for obtaining ice-thickness distribution for perennial sea ice or in stable thin sea-ice areas, it is still an unresolved issue for the seasonal sea-ice zone. In this study, we approach the problem of ice-thickness retrieval by using L-band synthetic aperture radar (SAR). In the SIZ, ice-thickness growth is closely related to the ridging activity and therefore surface roughness is expected to be correlated with ice thickness. L-band SAR is suitable for detecting such surface roughness and therefore is expected to be a good tool for obtaining thickness distribution. However, the available L-band SAR data have been very limited so far. To verify this idea, we conducted the measurements of ice thickness and surface roughness with the PV ‘Soya’, coordinated with airborne L-band SAR observations (horizontal resolution = 3 m) in February 2005 in the southern Sea of Okhotsk and confirmed that backscatter coefficients represent surface roughness well and resultantly have a good correlation with ice thickness. To further extend this result to ALOS/PALSAR whose horizontal resolution is 100 m, we conducted the first concurrent observation of ice thickness, surface roughness and satellite L-band SAR with the same ship in the same area in February 2008. The surface elevation was estimated with a heli-borne laser profiler, and ice thickness was measured by a ship-borne video monitoring system. To correct the slight difference in observation time, the effect of sea-ice drift was also taken into account. As a result, it was found that whereas it is difficult to discriminate sea ice from open water only from a single polarization SAR data, in the sea-ice area which was selected from video data backscatter coefficients are correlated well with ice thickness and surface roughness, defined by standard deviation of surface elevation. This result sheds light on the possibility of determining ice-thickness distribution in the SIZ. The regression obtained from this analysis allowed us to depict the ice-thickness distribution in the wide area. The results will also be discussed.