We have used crosscorrelation, semblance, and eigenstructure algorithms to
estimate coherency. The first two algorithms calculate coherency over a mul
tiplicity of trial time lags or dips, with the dip having the highest coher
ency corresponding to the local dip of the reflector. Partially because of
its greater computational cost, our original eigenstructure algorithm calcu
lated coherency along an implicitly flat horizon. Although generalizing the
eigenstructure algorithm to search over a range of test dips allowed us to
image coherency in the presence of steeply dipping structures, we were som
ewhat surprised that this generalization concomitant ly degenerated the qua
lity of the fault images in flatter dip areas.
Because it is a local estimation of reflector dip (including as few as five
traces), the multidip coherency estimate provides an algorithmically corre
ct, but interpretationally undesirable, estimate of the best apparent dip t
hat explained the offset reflectors across a fault, We ameliorate this prob
lem using two methods, both of which require the smoothing of a locally ina
ccurate estimate of regional dip. We then calculate our eigenstructure esti
mate of coherency only along the dip of the reflector, thereby providing ma
ximum lateral resolution of reflector discontinuities. We are thus both bet
ter able to explain the superior results obtained by our earliest eigenstru
cture analysis along interpreted horizon slices, yet able to extend this re
solution to steeply dipping reflectors on uninterpreted cubes of seismic da
ta.