This is in response to a post yesterday from Kate (I think it was Kate).
Here is a general summary of current theories on how storms can
affect distant (and local seismicity). This post is speculative and
contains back-of-the-envelop calculations which could be questioned, but
the general idea should not be abandoned due to the LOD and GPS
observations.
As you know, the earth's crust is made up of a series of large
plates. These plates move about the surface of the earth by being
pushed by spreading oceans at mid-ocean ridges and by their own
weight pulling them into the earth at subduction zones. A very small
amount of the motion is due to convection currents under the plates.
The movement of these plate is excrutiatingly slow. Motion on the
San Andreas is on the order of 2-3 cm per year. In one day, the
actual motion is therefore about 100th this - about the width of a
hair. In a second, the plates may move past each other about
the distance of 10 atoms set in a row.
But each plate moves as a unit, it does not distort as it
moves, or we would see distortion features within the plate (such
as mountains etc.). Therefore a force exerted on the plate in one
place will affect the motion of the entire plate.
Geologists are entirely concerned with "geology" - the rocks
that make up the earth. This prejudice has prevented many from
realizing that the plates actually have two surfaces - one at the
bottom of the plate and the second at the top of the plate. Each
of which can be operated on by external forces such as fluid drag.
At the bottom the plastic material has about the deforming (viscosity)
ability of a wrecking ball. At the top the air has the deforming
ability of a gas. At the bottom the motion of the plastic aesthenosphere
is measured in cm per year. At the top, the motion is measured in
miles (or km) per hour. Energy is transformed from kinetic energy
of these two masses to motion of the tectonic plate. Let's see
how much motion this is:
Kinetic energy is governed by the equation: K.E = 1/2mv*v - note
that the mass here is relatively minor compared with the velocity
of the mass. So, if we have a convection cell under the U.S.
moving at 3 cm/year (about 0.000003 m/hr or 0.0000000003 km/hr)
and this has a density of 5 g/cc over an interface the size
of the U.S. - about 9 million sq. km, the an approximation to
the kinetic energy imparted to the plate in a second by this
convection is about 1/2*5*9million*0.0000000003*0.0000000003
ignoring units which will be the same in the comparison to come
This number is 0.000000000002.
Now we can compare this with the energy from a storm surge on
the coast of Florida. The storm surge is moving at about 10 km/hour
with a density of 1 g/cc over an interface covering about 10km X 200KM -
about 2000 sq km. We can find the energy of this storm surge -
1/2*1*2000*10*10 = 100000 units of energy (units are the same
as in calcuation above). Even if we look only at the energy
imparted by the wind which is moving at 150 km/hr over an area of
100 km by 100 km - the density is much smaller however - around 0.001 g/cc
so the equivalent factor for wind energy transferred to kinetic
energy of the plate is about 1/2*0.001*10000*150*150 = 112,500 units
of energy, about the same as the storm surge.
But the important thing to note here is that the amount of kinetic
energy which can be transferred to the plate through landfall of
a hurricane is vastly greater than the kinetic energy transferred
to the plate in the same time (one second) by all the convective
forces operating on the bottom of the plate. This means that the
motion of the plate can easily be affect or even reversed temporarily
by hurricanes hitting the plate.
There is plenty of evidence that this does, in fact occur.
The time it takes the earth to rotate through a complete day has
been meticulously measured for decades. We know this period within
microseconds for each day - this is referred to as LOD (Length
of Day). Examination of this data has shown that the earth rotation
slows and speeds up in response mostly to weather phenomena such
as hurricanes.
Since hurricanes do not affect the ocean floor, the only area where
they would have a direct interaction with the earth would be where
they hit land. But what is the mechanism that causes the earth
to slow or speed up (depending on the direction the hurricane is moving)?
The hurricane must be imparting a substantial motion to the land
as it strikes. This in turn, moves the plate which affects the
motion of the earth on it's axis.
My first experience with the moving power of winds occurred many
years ago when I was hitch-hiking up the east African Rift valley and
found myself spending the day and evening in the ruins of Zimbabwe,
an ancient city in southern Africa. High above the city is the
Acropolis where the local people worshipped. On this day, I decided
to walk up to the Acropolis. Once there, a strong storm blew in.
After being hit by lightning and lying on the ground for minutes before
waking up, I sought a way off the mountain. The road had been washed
away, so I decided to take a ravine down the side of the mountain.
The ravine got narrower and narrower until the only way to move
down it was to press the two sides and chimney down. It was then
I realized the entire mountain was perceptibly shaking - you could
feel the mountain move with the wind.
If wind can move mountains, then they can move plates.
Since the plate moves as a whole, the motion imparted in Florida
is seen in California. During a recent spate of hurricanes several
years ago,GPS measurements showed an actual reversal of relative motion
between the North American Plate and the Pacific plate. This
was interpreted by some as a slow earthquake in Washington/Oregon,
but was observable throughout the entire western U.S. The plate
had changed it's motion, probably due to the hurricanes on the
eastern Seaboard. If the motion of the plate changes so does
the potential for earthquakes. If there is no relative motion,
then the possibility of earthquakes is considerably reduced. If
the motion increases, so should the seismic hazard.

I should mention that strong storm systems can also affect
seismicity locally. Pushing the N.American plate to the west,
as many hurricanes would do, tends to open up faults on the
mid-Atlantic ridge allowing stronger than normal earthquakes
to occur there. There is also some evidence that the low pressure
associated with hurricanes may reduce load and allow a "springing"
action to the land that results in an earthquake.

This is all speculative, but there seem to be sufficient
observations to work on the premise that an effect is possible
if not likely.