Fractals - Geomathematics: Theoretical Foundations, Applications and Future Developments - page 393

Fig. 10.18 Locations of
698 stream sediment
samples in the Iskut map
sheet, northwestern British
Columbia, part of National
Geochemical
Reconnaissance survey
conducted by Geological
Survey of Canada (Source:
Cheng et al. 1996 , Fig. 8)
precious-metal mineralization are commonly spatially related and may be genet-
ically associated with alkali-feldspar porphyry intrusions (Anderson 1993 ;
Cheng, 1 994). Sedimentary and volcanic clastic rocks of both the Stuhini and
Hazelton groups are favourable for Au mineralization in the area (Alldrick 1987 ;
Anderson 1989 , 1993 ;Cheng 1994 ).
In total, 698 stream sediment samples were collected from the area (Geological
Survey of Canada, GSC Open file 1645), providing coverage at a reasonably
uniform density, but irregularly distributed (Fig. 10.18 ). Samples were analyzed
for a suite of trace elements including Au, Ag, Cu, Mo, As, Sb and Pb. The data
were used to characterize some of the geochemically distinct rock units. Cheng
( 1994 ) showed that stream sediment elements Au, Ag, Cu, As and Sb are spatially
associated with Au mineralization. Gold content of the 698 samples is shown i n a
Q-Q plot (Fig. 10.19 ). In the present context, the goal is to delineate areas that are
anomalous with respect to stream sediment Au. Results of application of the C-A
method are shown in Fig. 10.20 . As in the previous example of gold in the Mitchell-
Sulphurets area, a model of two distinctly different straight-line segments provided
a good fit suggesting that gold anomalies are superimposed on background. The
study area is underlain by different rock types but this heterogeneity is not visible in
the Q - Q plot for gold (Fig. 10.19 ). Additional results for other elements and
anomaly maps are given in Cheng et al. ( 1996 ).
More recent publications in which the C-A method is applied for anomaly
detection include Lima et al. ( 2008 ), Carranza ( 2008 ), Park and Chi ( 2008 ) and
Zuo et al. ( 2009 ). The approach also has been extended from 2-D to 3-D (Afzal
et al. 2011 ; Wang et al. 2013 ).