Earthquake Report: Antarctic plate!

We just had an interesting mid-plate earthquake (not along a plate boundary). Hat Tip to Jascha Polet, who pointed out this is in the region of the 1998 M 8.1 earthquake, one of the largest strike-slip and mid-plate earthquakes ever recorded. I then learned that the seismicity in this region may be related to isostatic adjustments in the Antarctic plate! Here is the USGS website for this M 5.9 strike-slip earthquake.
Here is my interpretive map. I plot the USGS location as a yellow star. I also include some other figures as insets. I will discuss these below. I include a figure from Kreemer and Holt (2000) that shows focal mechanisms for earthquakes in the region plotted on a bathymetric map (seafloor topography). I also include a few maps from Das and Henry (2003). About a week ago, there was an earthquake along the Australia-Pacific plate boundary to the northeast of this earthquake (here is the Earthquake Report for that earthquake).
There is a legend that shows how moment tensors can be interpreted. Moment tensors are graphical solutions of seismic data that show two possible fault plane solutions. One must use local tectonics, along with other data, to be able to interpret which of the two possible solutions is correct. The legend shows how these two solutions are oriented for each example (Normal/Extensional, Thrust/Compressional, and Strike-Slip/Shear). There is more about moment tensors and focal mechanisms at the USGS.


Here is the earthquake report interpretive poster for the recent earthquake to the northeast.


Here is the Kreemer and Holt (2000) figure 1, showing the focal mechanisms for earthquakes along the regional plate boundary faults, as well as the focal mechanisms from the earthquakes in the region of the 1998 M 8.1 earthquake. I include their figure caption below in blockquote.

Focal mechanisms are from the Harvard CMT catalog (1/77-6/99). The black focal mechanisms indicate the 1998 Antarctic plate event with (some of) its aftershocks. Bathymetry is from Smith and Sandwell [1994]. Transform locations are derived from satellite altimetry by Spitzak and DeMets [1996]. MRC is the Macquarie Ridge Complex and TJ is the Australia-Pacific- Antarctica triple junction.

Here is the first figure from Das and Henry (2003). They plot the epicenters and focal mechanisms for earthquakes from the 1998 swarm overlain upon the gravity anomaly map. I include their figure caption below in blockquote.

The 25 March 1998 Antarctic plate earthquake (with a seismic moment of 1.3  1021 N m). (a) Relocated aftershocks [Henry et al., 2000] for the period 25 March 1998 to 25 March 1999 are shown as diamonds, with the main shock epicenter shown by a star. Only those earthquakes which are located with the semimajor axis of the 90% confidence ellipse 20 km are shown. International Seismological Centre epicenters for the period 1 January 1964 to 31 July 1997 are shown as circles. Marine gravity anomalies from an updated version of Sandwell and Smith [1997], illuminated from the east, with contours every 20 mGal, are shown in the background in the epicentral region. Selected linear gravity features are identified by white lines and are labeled F1–F6. F1, F2, and their southward continuation to join F1a compose the George V fracture zone. F4–F6 compose the Tasman fracture zone. (b) An expanded view of the region of the aftershocks. The relocated aftershocks in the first 24 hours are shown as diamonds; the rest are shown as circles. The 90% confidence ellipses are plotted for the locations; earthquakes without confidence ellipses were not successfully relocated and are plotted at the National Earthquake Information Center (NEIC) locations. The yellow star shows the NEIC epicenter for the main shock, with the CMT mechanism of solution 5 from Henry et al. [2000]. Available Harvard CMT solutions for the aftershocks are plotted, linked with lines to their centroid locations and then to their relocated epicenters, and are identified by their dates (mmddyy). The location of the linear features identified on Figure 6a are shown by black arrows. (c) Final distribution of moment release for preferred solution 8 of Henry et al. [2000]. There are the same gravity anomalies, same linear features, and same epicenters as Figure 6b except that now only earthquakes which are located with the semimajor axis of the 90% confidence ellipse 20 km are shown. Two isochrons from Mu¨ller et al. [1997] are plotted as white lines. Superimposed graph shows the final moment density, with a peak density of 1.25  1019 N m km 1. Regions of the fault with 15% of this maximum value are excluded in this plot. The baseline of the graph is the physical location of the fault. The spatial and temporal grid sizes used in the inversion for the slip were 5 km 5 km and 3 s, respectively.

This is the continuation of the above figure. This shows their interpretation of the faults that slipped during this 1998 earthquake series. In their paper, Das and Henry (2003) discuss the relations between main shocks and aftershocks. At the time, the 1998 earthquake “was the largest crustal submarine intraplate earthquake ever recorded, the largest strike-slip earthquake
since 1977, and at the time the fifth largest of any type worldwide since 1977” (Das and Henry, 2003). This M 8.1 earthquake was interesting because it crossed the fracture zones that trend N-S in the area. This is especially interesting because this is also what happened during the 2012 Sumatra Outer Rise earthquakes. Toda and Stein (2000) model the coulomb stress changes associated with different slip models from the M 8.1 earthquake to estimate if the aftershocks were triggered by the main earthquake. I include their figure caption below in blockquote.

(d) Principal features of the main shock rupture process [from Henry et al., 2000]. Arrows show location and directivity for the first and second subevents. Arrows are labeled with start and end times of rupture segments. Focal mechanisms are shown for the initiation, the first subevent plotted at the centroid obtained by Henry et al. [2000], and the second subevent. (The second subevent is not well located, and the centroid location is not indicated.) The cross shows the centroid location of moment tensor of the total earthquake obtained by Henry et al. [2000], and the triangle shows the Harvard CMT centroid. The same aftershock epicenters as Figure 6c are shown. Linear gravity features are shown as shaded lines, and probable locations of tectonic features T1a and T3a associated with the gravity features F1a and F3a are shown as shaded dashed lines. (See Henry et al. [2000] for further details.)

Here is the Kreemer and Holt (2000) figure that shows their interpretation of the stress field. The first figure below shows their determination of the strain rates as modeled from tectonic stresses at the plate boundaries. Note the low strain rate in the area near the M 8.1 earthquake (plotted as a focal mechanism). The second figure below shows the averaged minimum horiztonal deviatoric stress field caused by by flexure in the crust following the last ice age. Based upon their analyses, they attribute the earthquake to possibly be the result of stresses in the Antarctic plate following the last deglaciation. I include their figure caption below in blockquote.

a) Grid in which a strain rate field is determined associated with the accommodation of relative plate motions [DeMets et al., 1994]. These motions are applied as boundary velocity conditions,
illustrated by the grey arrows. b) Principal axes of the strain rate field for the region where the Antarctic event occurred (indicated by CMT focal mechanism). Model strain rates in this
region are one order of magnitude lower than along the surrounding ridges and transforms.


Principal axes of the vertically averaged minimum horizontal deviatoric stress field caused by gravitational potential energy differences within the lithosphere. CMT focal mechanism of Antarctic plate earthquake is shown. a) ‘ice-age’ simulation. b) change in stress tensor field from ‘ice-age’ to present day determined by taking the tensorial difference between the two solutions.

Leave a Reply

Your email address will not be published. Required fields are marked *