Page 109 - Geochemical Anomaly and Mineral Prospectivity Mapping in GIS

108 Chapter 4

the total variance of the stream sediment uni-element data explained by PC3 compared
to that explained by PC4 suggests that Cu-As anomalies are slightly more widely
distributed in the area than As-Ni anomalies. In addition, the magnitude of the loadings
of Cu and As on PC3 and PC4 suggests that the former is slightly mobile (thus more
dispersed) than the latter in the surficial environments of the study area. Based on these
arguments, it can hypothesised that, in terms of indicating presence of epithermal Au
mineralisation in the area, (a) the Cu-As association represented by PC3 are distal
anomalies whilst the As-Ni association represented by PC4 are proximal anomalies and,
thus, (b) the latter multi-element association (or PC4) is more important than the former
multi-element association (or PC3). Thus, the scores of PC3 and PC4 are further
subjected to the concentration-area fractal method for recognition of anomalies, although
results of analysis based on PC4 scores are explained first followed by results of analysis
based on PC3.
The scores of PC3 and PC4 for the point geochemical data are interpolated via
inverse distance moving average method to derive continuous geochemical surfaces. In
addition, the scores of PC3 and PC4 obtained here for the point geochemical data are
attributed to pixels in the associated stream sediment sample catchment basins to derive
discrete geochemical surfaces. The multi-element geochemical surfaces are discretised in
the same way the uni-element geochemical surfaces are discretised (see above). The
plots of concentration-area relations for the multi-element geochemical surfaces are
shown in Fig. 4-17. Note that the ‘concentration’ variables represented by the PC scores
do not have the normal concentration units because the PCs are linear combinations of
the log e-transformed uni-element data. For this reason and because negative PC scores
cannot be transformed to logarithms, the ‘concentration’ axes of the concentration-area
plots are not in the logarithmic scale. The PC scores at the breaks in slopes of the straight
lines fitted to the concentration-area relations represent thresholds that can be used to
classify the multi-element association scores into background and anomalous
populations. The very similar shapes of the concentration-area curves and the equal
numbers of thresholds defined per set of PC scores represented as continuous and
discrete geochemical surfaces (Fig. 4-17) suggest that, in this case study, either
continuous geochemical surfaces or discrete geochemical surfaces can be used in the
concentration-area fractal analysis of geochemical anomalies.
For the PC4 scores, the thresholds based on analysis of either continuous or discrete
geochemical surfaces indicate five populations, which are interpreted, from lowest to
highest, as (a) low background, (b) background, (c) high background, (d) low anomaly
and (e) high anomaly. The spatial distributions of the background and anomalous
populations of PC4 scores, representing As-Ni association in stream sediments, show
some degree of similarity (Fig. 4-18). For the PC3 scores, the thresholds based on
analysis of either continuous or discrete geochemical surfaces indicate four populations,
which are interpreted, from lowest to highest, as (a) low background, (b) high
background, (c) low anomaly and (d) high anomaly. The spatial distributions of the
background and anomalous populations of PC3 scores represented as continuous and