An earthquake of Mw 6.2 occurred in Kumamoto Prefecture, Kyushu Island, Japan at 21:26 on April14, 2016. About 28 hours after, another earthquake of Mw 7.0 occurred at 1:25 on April 16. Surface ruptures withlateral displacement of up to 2 m appeared along the Futagawa fault line. A large number of buildings andinfrastructures were severely damaged. For a prompt emergency response, it is important to grasp damage distributionat an early stage after the occurrence of an earthquake. The purpose of this research is to extract landslides andgeological effects caused by the Kumamoto earthquake using airborne Lidar data. Digital surface models (DSMs)acquired before and after the April 16 earthquake were employed to grasp slope failures in Mashiki town. A methodto extract landslides automatically was investigated by differentiating the DSMs at two time instants and thecharacteristics of the DSM difference in the area where landslides occurred were examined. Then the result ofautomatic extraction was compared with those from visual interpretation of aerial photographs. From this comparison,it is concluded that small landslides, which are difficult to locate from field survey or aerial-photo interpretation, aswell as large-scale ones, were extracted successfully from the DSM difference.