Date: Fri, 21 Jan 2005 10:18:20 -0800
From: Richard Gross 

Hello Ron -

By changing the mass distribution of the Earth, and hence its inertia 
tensor, conservation of angular momentum dictates that the Sumatran 
earthquake of December 26, 2004 should have changed the rotation of the 
Earth. Models of the effect of the earthquake on the Earth's rotation 
predict that the length of the day should have become shorter by 2.68 
microseconds and that the figure axis of the Earth should have shifted by 
0.821 milliarcseconds in the direction of 145 East longitude (Chao and 
Gross, "Did the 26 December 2004 Sumatra, Indonesia, Earthquake Disrupt 
the Earth's Rotation as the Mass Media Have Said?", EOS, vol. 86, No. 1, 
pp. 1-2, 2005). 

Space-geodetic measurement techniques like satellite laser ranging (SLR) 
are able to directly estimate changes in the length-of-day, but not 
changes in the figure axis; instead, SLR directly estimates the position 
of the rotation axis. Of course, the motion of the rotation axis will 
change in response to the shift of the figure axis, and it is this 
response of the rotation axis to the shift in the figure axis caused by 
the earthquake that may have been detected in SLR measurements.

However, the figure axis, and hence the rotation axis, will change in 
response to any change in the mass distribution of the Earth, including 
changes in the mass distribution of the atmosphere and oceans. In fact, 
these changes in the mass distribution of the atmosphere and oceans are 
nearly 100 times larger than the changes predicted to have been caused by 
the earthquake. Prior to attributing a change in the position of the 
rotation axis to the Sumatran earthquake, I think that the much larger 
changes caused by the atmosphere and oceans should first be accounted for. 
Since the atmosphere and oceans are in constant motion, they will also 
affect the motion of the rotation axis in December, and the particular 
change attributed to the earthquake may instead be caused by the 
atmosphere and/or oceans. 

While the change in the motion of the rotation axis occurs near the time 
of the earthquake, similar shifts in the motion of the rotation axis can 
be observed at other epochs. For example, global positioning system (GPS) 
estimates show the same change in the motion of the rotation axis near the 
time of the earthquake (igscb.jpl.nasa.gov/images/Dec2004EQ.Gendt.gif).
However, the GPS estimates show that a similar change in the motion of the 
rotation axis, albeit of smaller amplitude, also occurred near December 19, 2004.

So, in summary, without accounting for the larger effects of the 
atmosphere and oceans on the motion of the rotation axis, I think it is 
premature to attribute the detected change to the earthquake.

The change in the position of the figure axis caused by the earthquake 
should have a unique signature --- it should appear as a step-like change. 
The earthquake has permanently redistributed the Earth's mass, causing a 
permanent change in the length-of-day and in the position of the figure 
axis. Since the atmosphere and oceans are fluids, they cannot support a 
permanent redistribution of mass; hence, the changes caused by the 
atmosphere and oceans will not be step-like. They can still cause sudden 
changes in the position of the figure axis since the atmospheric and 
oceanic mass can move suddenly, but the atmospheric and oceanic mass 
should then move back again with no net, permanent, redistribution of 
mass. So, to unambiguously attribute a change in the Earth's rotation to 
the earthquake, a step-like change in the position of the figure axis at 
the time of the earthquake should be looked for. I will, of course, be 
doing this as soon as estimates of the atmospheric and oceanic angular 
momentum for December and January become available. Removing the effects 
of the atmosphere and oceans on the figure axis should make it easier to 
detect a step-like change.

Best regards,
Richard