2-D and 3-D Trend Analysis - Geomathematics: Theoretical Foundations, Applications and Future Developments - page 256

Fig. 7.16 Contours (in feet above sealevel) of parts of eight quadratic trend surfaces fitted to
bottom of Matinenda Formation, Elliot Lake area (Source: Agterberg 1984 , Fig. 8)
Fig. 7.17 Sum of trend (see Fig. 7.16 ) and signal (see Fig. 7.15 ) for bottom of Matinenda
Formation. Note strong resemblance between patterns of Figs. 7.16 and 7.17 illustrating that the
fluctuations (signal) contoured in Fig. 7.15 have relatively small magnitudes (Source: Agterberg
1984 , Fig. 9)
equation), “signal” (characterized by a homogeneous autocorrelation function), and
stochastically independent “noise” have different statistical connotations, the dif-
ferences between the features they represent usually are less distinct. It will be seen
that some of the signal in the present example represents structural deformation. It
also is known that these concepts depend on the sampling density. In general, if
more measurements are performed, some of the signal becomes trend and some of
the noise becomes signal.
Although it can be assumed that the major uranium-producing conglomerates in
the Elliot Lake area were deposited in channels, the geometrical pattern of the
channels remains a subject of speculation, at least in places removed from the
mining areas. Bain ( 1960 ) assumed a single “uraniferous river channel” (see also
Stanton 1972 , Figs. 12-16) winding its way through the basin so that it fits the
approximately NW trending channels at the Quirke, Nordic and Pronto mineralized