... recording^1.1

This process of combining several input information signals into one output signal is known as multiplexing in the communications industry. Recovering the individual signals from the multiplex signal is known as demultiplexing.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... operation^1.2

This definition is adjusted from that in http://en.wikipedia.org/wiki/Decoder.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... outcome^2.1

in MATLAB notation; likewise for the following arrays in this section

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... times^2.2

$\epsilon$ is a small fraction, for instance, $1/12$ .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... size^2.3

measured in the number of complex numbers

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... updates^2.4

Since the Hessian of the objective function is constant given a fixed ${\textrm{CSG}_\textrm{enc}}$ , those $K_{CGit}$ iterations are made using CG.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... factor^2.5

As $K_{CGit}$ increases, $K_{it}$ may also have to increase in order to produce acceptable result. Therefore this reduction factor is a bit smaller than $K_{CGit}$ .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... size^2.6

$n_x$ is reduced from the original value of 645 to speed up the FFT.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... function^3.1

The misfit function $\epsilon=\frac{1}{2}\vert\vert{\bf d}^{obs.}-{\bf d}^{ref.}\vert\vert^2$ relates to the $L_2$ norm of the encoded difference between the predicted and observed supergathers.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... finite-difference^3.2

A finite-difference simulation of two $simultaneous$ sources (a red source and a dark blue source) will compute traces everywhere on the surface that are a superposition of the wavefields from both sources.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...fig:2nt^3.3

Only a single $\delta(t)$ -impulse of the Earth response is shown. Linear superposition generalizes this to an arbitrary impulse response within $nt$ .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... smearing^4.1

The seismic amplitude is smeared over the thick ellipse shown in Figure 4.3a, where the period $T$ of the trace's source wavelet determines the thickness of the fat ellipse in $(x,z)$ space; Figure 4.3b illustrates that the minimum thickness of the fat ellipse as $0.5 \lambda$ .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... residual^4.2

The residual can be either the traveltime residual or the waveform residual.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... scatterer^4.3

We will assume a 2D model where the ``point'' source and scatterer are equivalent to a line source and a line scatterer, with no field variations along the y-axis.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... wavepath^4.4

Dahlen (2004) refers to the shape of a diving wavepath as a banana.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... wavepaths^4.5

There are two steps for creating an upgoing reflection wavepath: first, generate the migration image and use the reflectors as exploding sources that explode at the traveltime from the source to the reflector. Then, fire off these exploding reflectors to get the upgoing reflection fields $U({\bf{x}},t)$ . The upgoing rabbit ear wavepath is computed by taking the zero-lag correlation between $U({\bf{x}},t)$ and the backpropagated data $B({\bf{x}},t)$ .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... bed^4.6

The sampling interval between wavenumbers associated with each order of multiple becomes smaller with thinner beds.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... reflection^4.7

This kernel corresponds to just one of the terms in the Neummann series expansion of the Lippmann-Schwinger equation (Stolt and Benson, 1986).

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... waves^11.1

We exclude the case where the scatterer-diving wave interaction produces significant diffractions, so that all source-geophone pairs see significant diffraction energy, not just changes in the diving-wave arrival. This would be the case where the scatterer only has a velocity contrast but no impedance contrast.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.