PSN-L Email List Message

Subject: "Giant Seismometer"
From: John Hernlund hernlund@............
Date: Tue, 26 Dec 2000 16:12:48 -0800

Hi Everyone,
   I came across some really good stuff at AGU last weekend.  One
item that will interest you all is the construction of what may be the
largest seismometer ever.  A friend of mine named Steve Gao (at
Kansas State) got together with a fellow engineering professor at
KSU that can measure the length of fiber optic cable very accurately
using a simple reflection technique (the technology is in the sending
and receiving unit).  They got an NSF grant to go around hooking this
thing up to existing fiber optic networks and monitoring the strain
across the fiber optic cable in time.  They haven't yet conducted their
first major experiment, but they expect to measure a lot of subtle
effects.  The advantage of this device is that it is truly broadband,
and the technique can conceivably sample at any frequency without
any frequency dependence.  The sending/receiving unit costs around
$10,000 to build, which is cheaper than many commercial broadband
seismometers.  Since most of the fiber optic cables in the ground for
telecommunications are not in use (due to super-high bandwidth) they
can go around renting an optic cable here and there.  Their biggest
ambition is to wire up a huge network of these things, especially in
the area of the San Andreas where strain precursors may be looked
into for a very low price.  Traditional strain measurement at the San
Andreas has consisted of laser interferometry through the air, which
turns out noisy/messy data and is extremely expensive.  Plus the
desired accuracy can be achieved in the fiber by simply increasing the
length of the cable you use, since the error over time
will be independent of the fiber length.  If the instrument yields an
error of X (for example), and you wish to measure a strain above a
noise level E, then just hook it up to a cable of length L = X/E.  For
them, X is typically less than 1 mm, so for a 10 km cable the lower
limit of effective measurement is 10^(-7), which gives them a very
useful range.  Over time these things will become more accurate and
less expensive to use...


John Hernlund
Department of Earth and Space Sciences
University of California, Los Angeles


Public Seismic Network Mailing List (PSN-L)

[ Top ] [ Back ] [ Home Page ]

Larry Cochrane <cochrane@..............>