2000.07.23
Not-so-deep thoughts on seismic imaging and data processing:
In wave propagation theory, the seismic wavefield is usually described by two
complimentary representations: wavefront snapshots and raypaths. Neither description
matches the form in which the wavefield is observed during a seismic experiment.
Both the snapshot and the raypath representations assume knowledge of the complete
history of the wavefield in the subsurface, as well as knowledge of the subsurface
structures through which the wavefield passed.
In seismic exploration, the passage of a wavefield with all its component wavefronts
is detected as seismograms at a limited number of discrete locations, usually at the
earth's surface. During seismic data processing, the seismograms that compose a
reflection survey are used to infer the nature of the subsurface. Because a
seismogram is a transient signal, it is possible to study segments, or time windows,
of a seismic trace nearly independently of the whole trace. For an impulsive source,
individual wavelets in the recorded signal can be interpreted as events corresponding
to particular reflection or refraction travel paths. The travel paths cannot be
determined uniquely from a single shot record, but can be greatly constrained by
analysis of a large volume of redundant data acquired by thousands of shots.
(Redundant data also helps compensate for the low signal-to-noise quality typical
of seismic data, of course.)
O.k., so I'm writing this encyclopedia article on exploration seismology, see, and it's
basically been this big brain-dump, but I realized that some of this stuff was just
over-kill and it was really dragging things down. But while I'm writing I tend to get a
little emotionally attached to anything I commit to a word processor, and I have trouble
throwing away my own stuff. So I started thinking, "You know, I haven't done a
journal update in a few days ..."
2000.07.05
In a free market economy, it's hard to
prove price gouging of a commodity item. (WSJ subscription required.) In a
nutshell, natural market forces set the price of gasoline. Duh.
Of course, severe gasoline shortages really would be disruptive to the economy, and
could even constitute a threat to our national security. If we as a country were
faced with a problem of that magnitude, we would need think about reining in this
free market thing, and adopting an energy policy that injected a little regulation
into the oil and gas industry.
Such an energy policy would need to simultaneously guarantee a stable
supply and an adequate profit margin for gasoline. Naturally, that would be long-term
expensive. And I'm probably talking about more than two dollars a gallon expensive.
But relax, it's not going to happen.
