Page 199 - Introduction to Mineral Exploration

182 C.J. MOON & M.K.G. WHATELEY

TABLE 9.3 Part of the data associated with Fig. 9.3 stored as a flat file. Note that a change (e.g. name of
geological unit) would require editing of all instances.

Area Perimeter Modgeol_ Unit Lithology Age
12,983 711 2 Basic volcanics basic volcanics Middle Devonian
2648,177 8049 3 Upper Devonian slates slates Upper Devonian
14,476 866 4 Basic volcanics basic volcanics Middle Devonian
977,190 10,015 5 Sea none None
7062,567 27,324 6 Upper Devonian slates slates Upper Devonian
69,818,680 231,598 7 Mid Devonian slates slates Middle Devonian
16,556 598 8 Dolerite dolerite Upper Devonian
11,447,985 51,973 9 Basic volcanics basic volcanics Middle Devonian
415,067 5493 10 Mid Devonian slates slates Middle Devonian
8805 566 11 Basic volcanics basic volcanics Middle Devonian
6650 468 12 Mid Devonian limestone limestone Middle Devonian
860,230 5309 13 Basic volcanics basic volcanics Middle Devonian
9946 437 14 Dolerite dolerite Upper Devonian
47,900 958 15 Dolerite dolerite Upper Devonian
36,817 980 16 Dolerite dolerite Upper Devonian
15,053 492 17 Dolerite dolerite Upper Devonian
297,739 3253 18 Basic volcanics basic volcanics Middle Devonian
39,071 972 19 Dolerite dolerite Upper Devonian
156 86 20 Sea none None

format can be converted to a relational database as the database stores the objects directly, com-
by a process known as normalization. Table 9.4 plete with topological and other information
shows an example of this process, a unit num- links. For this reason, OO database systems are
ber (finalgeol) which is a simplified version being increasing used in Internet information
of modgeol, lithology number (lith#) and age# delivery where complex multimedia objects
have been added as the first step in this process. need to be retrieved and displayed rapidly
The next step is (Table 9.5) to break Table 9.3 (FracSIS 2002).
into its components. The first table of Table 9.5
link the polygons with a finalgeol. The second
links finalgeol and with lith# and age# (Table 9.1.4 Corporate solutions
9.5b). The third part of Table 9.5 links lith# As large amounts of money are invested in col-
with lithology name (Table 9.5c) and the fourth lecting the data, it is crucial that the data are
age# with age (Table 9.5d). This decomposition safely archived and made available to those
into their normalized form allows easy editing, who need them as easily as possible. Integrity
e.g. an error in the formation name can be of data is paramount for any mining or explora-
corrected with a single edit. tion company, both from a technical and
The flat file and relational database systems legal viewpoint (acQuire 2004). However this
are mature technologies and used extensively integrity has often been lacking in the past and
for 2D GIS and geometric models. It is however many organizations have had poor systems
becoming increasingly difficult for them to giving rise to inconsistencies, lost data, and
manage 3D information with complex topo- errors. Increasingly, in the wake of incidents
logical relationships. Difficulties arise in con- such as the Bre-X fraud (see section 5.4), both
verting the complex data types used in industry and government departments require
portraying these objects being represented into higher levels of reporting standards. Relational
relational tables with links. This problem is databases provide the means by which data can
overcome in an object-oriented (OO) database be stored with correct quality control pro-
approach in which conversion is unnecessary cedures and retrieved in a secure environment.