First observations of the seiche that shook the world

Abstract

On September 16th, 2023, an anomalous 10.88 mHz seismic signal was observed globally, persisting for 9 days. One month later an identical signal appeared, lasting for another week. Several studies have theorized that these signals were produced by seiches which formed after two landslide generated mega-tsunamis in an East-Greenland fjord. This theory is supported by seismic inversions, and analytical and numerical modeling, but no direct observations have been made -- until now. Using data from the new Surface Water Ocean Topography mission, we present the first observations of this phenomenon. By ruling out other oceanographic processes, we validate the seiche theory of previous authors and independently estimate its initial amplitude at 7.9 m using Bayesian machine learning and seismic data. This study demonstrates the value of satellite altimetry for studying extreme events, while also highlighting the need for specialized methods to address the altimetric data's limitations, namely temporal sparsity. These data and approaches will help in understanding future unseen extremes driven by climate change.

Figures (9)

• Figure 1: Dickson Fjord study region, SWOT measurements, and in-situ measurements (A) Sentinel-2 image of the Dickson Fjord in summertime, with rock-slide, CTD tide-gauge, and atmospheric station shown. (B) Visualization of the study region, and nearby IU.SFJD and II.ALE seismic station. Rayleigh and Love nodes are plotted in blue and orange. (C) SWOT pixelcloud measurements from a single pass over Dickson Fjord, measurements colored by measured sea-surface height. (D.i & D.ii) Seismic observations at II.ALE (green) scripps1986global and IU.SFJD (magenta) albuquerque1988global bandpass filtered between 10-13mHz for the September 16th (i) and October 11th (ii) events. Time units are minutes.
• Figure 2: Pixelcloud sea-surface elevation maps of the Dickson Fjord in the days following the two tsunamis. (A) SWOT observation of the fjord 0.5 days after the October 11th tsunami. Rayleigh and Love nodes are overlaid to show the theorized axis of propagation. (B,C,D) Consecutive SWOT observations of the fjord 0.5 days, 1.5 days, and 4.8 days after the September 16th event respectively.
• Figure 3: Seismic observations of September and October VLP signals at II.ALE Seismic Station and SWOT cross-channel slopes. (A,B) Normalized VLP signals filtered between 10-13 for the September and October events respectively. SWOT observations are given by vertical lines. (C-F) Normalized VLP signals with signals observed by SWOT shown as vertical lines. Observed magnitudes relative to the maximum amplitude are shown. (H-K) Corresponding SWOT cross-channel observations from $X_1$ to $X_2$. Slope estimates and associated Bayesian $R^2$ values from a Bayesian linear model are provided (Section \ref{['Bayes']}).
• Figure 4: Estimates of the dominant lunar tide, M2, from SWOT pixelcloud data using a spatially coherent variational Bayesian harmonic analysis. (Top) Estimated amplitudes, and (Bottom) corresponding phase lags. Estimates are only made for points which have at least 15 measurements. Inset plots show how the amplitude and phase vary along the rayleigh (blue, left to right) and love (orange, bottom to top) nodes respectively. The M2 amplitude and phase lag computed from the depth measurements at the CTD station are shown by the square.
• Figure 5: Wind speed and direction from the Dickson Fjord CTD station during each event(Top) Wind speed and direction over the duration of the September 16th VLP signal. (Bottom) Wind speed and direction over the duration of the October 11th VLP signal.