Predictive Modeling of Mineral Exploration Targets 11
Traditionally, modeling and mapping of significant geochemical anomalies were
based mostly on geochemical data sets, probably because in the past (say, before the
1970s when the development of GIS was at its infancy) many outcropping mineral
deposits were still undiscovered such that their associated geochemical anomalies are
obvious in the data sets. However, not all geochemical anomalies indicate the presence
of mineral deposits. Mineral deposits are themselves significant geochemical anomalies.
In more recent times, it can be surmised that most, if not all, outcropping mineral
deposits have already been discovered. In addition, mining and other industries have
already altered the geochemical landscapes. Therefore, an obvious geochemical anomaly
may be geogenic and significant (i.e., associated with mineral deposits), geogenic but
non-significant (i.e., related to certain high background non-mineralised lithologies), or
anthropogenic and non-significant (e.g., due to industrial contamination). In contrast, a
subtle geochemical anomaly or absence of a geochemical anomaly does not necessarily
indicate absence of mineral deposits, but may suggest either that weathering and erosion
rates were insufficient to mobilise and disperse metals from mineralised sources or that
mineral deposits are ‘blind’ or ‘buried’ (i.e., non-outcropping or unexposed to the
surface). It is clear, therefore, that effective modeling of significant geochemical
anomalies requires that all available relevant exploration data sets or pieces of geo-
information are analysed and integrated in the light of fundamental or theoretical
principles of exploration and environmental geochemistry.
Many modern methods for modeling of significant geochemical anomalies
subsequently consider not only geochemical data frequency distributions but also other
types of geo-information from uni-element or multi-element geochemical data sets (e.g.,
Grunsky and Agterberg, 1992; Bellehumeur et al., 1994; Cheng et al., 1996, 1997, 2000;
Cheng, 1999b; Harris et al., 2001a; Karger and Sandomirsky, 2001), namely: spatial
variability and correlations; geometry (shape and orientation, as well as fractal
dimensions) of anomalies; and scale independence of anomalies. Consideration of the
scale independence of anomalies aims to reduce the effects of sampling density and geo-
analytical techniques on the spatial distributions of geochemical anomalies.
Considerations of the spatial variability and correlations and the geochemical properties
of geochemical anomalies aim to enhance anomaly patterns that reflect controls by
geological processes and thus facilitate recognition of significant geochemical
anomalies. Enhancement of geochemical anomaly patterns that reflect controls by
geological processes can also be addressed by integrating geochemical data with other
types of relevant spatial data or pieces of spatial geo-information that explicitly represent
individual processes. For example, significant drainage geochemical anomalies can be
enhanced and thus recognised by integrating geochemical data with area of drainage
catchment basins (Polikarpochkin, 1971; Hawkes, 1976; Moon, 1999), lithologic units
(Rose et al., 1970; Bonham-Carter and Goodfellow, 1986; Bonham-Carter et al., 1987),
proximity to faults (Carranza and Hale, 1997); drainage sinuosity (Seoane and De Barros
Silva, 1999), and stream order (Carranza, 2004a). Significant soil geochemical
anomalies can be enhanced and thus recognised by integrating soil geochemical data
chiefly with lithology (e.g., Lombard et al., 1999; Garrett and Lalor, 2005; Jordan et al.,