Interpreting seismic amplitudes - 3-D Seismic Interpretation - page 139

Intercept*Gradient

Positive AVO

Negative AVO

Fig. 5.17 Intercept ∗ gradient section. Positive values are displayed in red and indicate pay sand.

be calculated for a range of incidence angles and stacked to give a trace that can be
compared with the real seismic data. In practice the match from seismic to zero-offset
synthetic is often reasonable for class I or class III reflectors. It is with class IIp events
that the worst problems arise, as the stacked amplitude may be very small and perhaps
opposite in sign to the zero-offset response. A poor tie to the zero-offset synthetic is
then inevitable.
A complication for AVO analysis is the effect of tuning. Because the TWT difference
between two closely spaced reflectors in a gather will vary with incidence angle (before
NMO correction), it follows that tuning effects will vary across the gather and distort
the AVO response in ways that are hard to recognise except close to a well where bed
thicknesses are known and tuning can be modelled.

5.5 Rock physics for seismic modelling

To understand observed amplitude effects, we often need to know how rock densities
and seismic velocities (both P and S) are affected by fluid fill (brine, oil or gas), by
porosity, by pressure, by clay content, and so on. This is a large subject. A detailed
account can be found in Mavko et al . (1998) , for example, and we shall only provide a
summary here. The methods we shall describe work best for medium- to high-porosity
sandstones, and are applicable to carbonates only when the pore structure is relatively
uniform with a pore size very much smaller than the sonic wavelength. Problems arise
in the case of rocks having low porosity and permeability; velocities recorded by the
sonic log may be different from those applicable to surface seismic data, because of
dependence of seismic velocity on frequency ( dispersion ).