The second highest tsunami ever recorded

Over the past century, more than two dozen tsunamis with runups greater than 50 m have been reported. Of these, more than half were in the Arctic or subarctic, including the 1958 Lituya Bay megatsunami, which ran up 530 m in elevation. Many of those megatsunamis occurred in deglaciating fjords or valleys, and almost all were triggered by landslides. At 5:26 a.m. local time on 10 August 2025, a large landslide (>64x106 m3) collapsed about one vertical km onto South Sawyer Glacier and into Tracy Arm, a cruise ship-frequented fjord in southeast Alaska. The landslide triggered a megatsunami, which reached an elevation of 481 m up the opposite fjord wall before propagating out of the fjord into Stephens Passage and Endicott Arm to the south. The tsunami was experienced by multiple ships in the vicinity, but due in part to its early morning timing, luckily no deaths or injuries occurred. The initial rock wedge failure transitioned into a rock avalanche as it traveled down the slope and produced globally observed long-period seismic waves equivalent in size to those of a M5.4 earthquake. The landslide was also preceded by more than 24 hours of microearthquakes attributed to slip along the failure surface, with increasing rate and amplitude until roughly an hour before failure. Long-period monochromatic global seismic signals persisted for over 36 hours. These signals are consistent with a landslide-induced seiche trapped in the fjord, an interpretation further confirmed by tsunami simulations and satellite observations. The landslide and resulting hazard cascade was enabled by >6 km of retreat of South Sawyer Glacier since 1948, which we statistically attribute to anthropogenic warming. Most recently, the base of the failed slope was fully exposed between late June and early August 2025 when the glacier retreated several hundred metres. A large failure of a slope that had not previously been identified as a hazard, and with no precursory slope deformation, in a fjord with extensive recreational activity (e.g., cruise ships and personal pleasure craft) highlights the near-field risk of landslide tsunamis and underscores the importance of enhanced monitoring and continued research using a variety of tools, including seismic, remote sensing, and in situ monitoring.