Carl Tape's website

Recording the Aurora at Seismometers Across Alaska

Carl Tape, Adam Ringler, Don Hampton

Seismological Research Letters, 2020

Across Alaska there are six all-sky cameras, 13 magnetometers, and >200 seismometers. The all-sky images and magnetometers have the same objective, which is to monitor space weather and improve our understanding of auroral activity, including the influence on magnetic fields in the ground. These variations in the magnetic field are also visible on seismometers, to the extent that during an auroral event, the long-period (40–800 s) waves recorded by a seismometer are magnetic field variations, not true ground motion. Although this is a problem—one that can be rectified with magnetic shielding at each seismometer site—it is also an opportunity because the present seismic array in Alaska is much broader than the coverage by magnetometers and all-sky cameras. Here we focus on three aurora events and document a direct link between aurora images in the night sky and seismometer recordings on ground. We also dive into the history of scientific discoveries in this realm, starting in 1716.

For a public talk on some of these results, please check out our 2021 Science For Alaska talk.

Cover of November 2020 Seismological Research Letters issue, featuring a photograph by Aaron Lojewski that appears in Figure 2a of Tape, Ringler, Hampton (2020).

Weekly news in Science magazine (August 7, 2020) featuring the paper by Tape, Ringler, Hampton (2020) and the photograph by Aaron Lojewski.

Comparison between optical imagery and seismic data for 28 February 2019 UTC. (a) Seismic data from POKR (Poker Flat Research Range, Alaska) band-pass filtered from 100 to 300 s. (b) Normalized intensity of the red, blue, and green components of the all-sky camera at Poker Flat. The intensities are normalized to the peak value, which occurred on the green band at about 10:00 UTC. The envelope of the vertical component from (a) is shown in gray. The steep rise from the camera traces at the beginning and end are from scattered sunlight from twilight. [Figure 5 from Tape, Ringler, Hampton (2020).]

Sequence of events leading to the detection of natural magnetic variations on seismometers. (a) Long-period (>100 s), vertical-component seismometer recording from Stuttgart (station STU) from 2 to 8 May 1992. The expected oscillatory pattern is from the Earth tides. During the strike of electric train workers, the "fuzzy" overlying noise is absent. Figure courtesy of Erhard Wielandt. (b) The strike motivated the design, by Wielandt, of an active electromagnetic shielding "cube" for STU (Forbriger, 2012). Photograph by Thomas Forbriger. (c) Seismometer recordings from 21 February 1994 showing a magnetic storm caused by coronal mass ejection and starting at 09:01 UTC (Petrinec et al., 1995). Seismograms are from BFO (Black Forest Observatory, Germany) and COL (Fairbanks, Alaska), east component, filtered 40–800 s. The magnetic storm signal was suppressed at STU because of its active magnetic shielding (Klinge et al., 2002). [Figure A3 of Tape, Ringler, Hampton (2020)]

Figure from 1716 paper by Edmund Halley, who correctly depicted the Earth as having a magnetic field similar to a dipole. Halley's paper followed an aurora display seen across Europe in March of 1716.