Co-ordinate systems & projections

Geologists have in the past generally managed

to avoid dealing with different coordinate

systems in any detail, as the areas they were

dealing with were small. The advent of GPS

and computerized data management has

changed this. The plotting of real world data on

a flat surface is known as projection and is the

result of the need to visualize data as a flat

surface when the shape of the earth is best

approximated by a spheroid, a flattened sphere.

For small areas the distortion is not important

but for larger areas there will be a compromise

between preserving area and distance relationships.

For example, the well-known Mercator

projection emphasizes Europe at the expense of

Africa. The scale of the data also governs the

choice of projection. For maps of scales larger

than 1:250,000, either a national grid or a

Universal Transverse Mercator (UTM) grid is

generally used. In the latter projection, the

earth is divided into segments of 6 degrees

longitude with a value of 500,000 m E given to

the central meridian of longitude and a northing

origin of 0 m at the equator, if north of the

equator, or large number, often 10,000,000 m,

if south of the equator.

There are a variety of different values in use

for the ellipsoid that approximates the shape

of the earth, known as the datum. The most

commonly used datum for GPS work is World

Geodetic System (WGS 1984) but the datum

used on the map must be carefully checked,

as the use of different datums can change coordinates

by up to 1500 m. The reader is advised

to read about the problems in more detail in

texts such as Longley et al. (2001) and Snyder

(1987).

Corporate solutions

As large amounts of money are invested in collecting

the data, it is crucial that the data are

safely archived and made available to those

who need them as easily as possible. Integrity

of data is paramount for any mining or exploration

company, both from a technical and

legal viewpoint (acQuire 2004). However this

integrity has often been lacking in the past and

many organizations have had poor systems

giving rise to inconsistencies, lost data, and

errors. Increasingly, in the wake of incidents

such as the Bre-X fraud (see section 5.4), both

industry and government departments require

higher levels of reporting standards. Relational

databases provide the means by which data can

be stored with correct quality control procedures

and retrieved in a secure environment.

Many proprietary technical software products

provide such storage facilities, for example

acQuire (acQuire 2004) provides such a solution

for storage and reporting of data that also

interfaces with files in text formats such as csv,

dif, txt (tab delimited and fixed width formats),

as well as numerous proprietary formats.

The strategy for collection and evaluation

(checking) of data (Walters 1999) is often a matter

of company procedure. Most errors are gross−90

and can be easily filtered out. Each geologist

and mining or processing engineer knows what

the database should contain in terms of ranges,

values, and units. It is a simple matter of setting

up the validation tables to check that the

data conform to the ranges, values, and units

expected. A simple example would be ensuring

that the dip of drillholes is between 0 and

degrees for surface drilling.

Data Capture & Storage

There are two major methods of representing

spatial data, raster and vector. In the vector

model the spatial element of the data is represented

by a series of coordinates, whereas in the

raster model space is divided into regular pixels,

usually square. Each model has advantages and

disadvantages summarized but the

key factors in deciding on a format are resolution

and amount of storage required. A simple geological map in vector and

raster format. The raster method is commonly

used for remote sensing and discussed, whereas the vector method is used

for drillholes and geological mapping. Most

modern systems allow for integration of the

two different types as well as conversion from

one model to another, although raster to vector

conversion is much more difficult than that

from vector to raster.

In a simple (two-dimensional) vector model,

points are represented by

lines as a series of connected points (known as

vertices), and polygons as a series of connected

lines or strings. This simple model for polygons

is known as a spaghetti model and is that

adopted by computer aided drawing (CAD)

packages. For more complex querying and

modeling of polygons, the relationship between

adjoining polygons must be established and

the entire space of the study area subdivided.

This is known as the topological model. In this

model, polygons are formed by the use of

software as a mesh of lines, often known as

arcs, that meet at nodes. Another variation of

this model often used for height data is that of

the triangular irregular networks (TIN) and is

used to visualize digital elevation surfaces or

construct digital terrain models (DTM). The

TIN model is similar to the polygons used in

ore resource and reserve calculationx and y coordinates,

Introduction

One of the major developments in mineral

exploration has been the increased use of computerized

data management. This has been used

to handle the flow of the large amounts of data

generated by modern instrumentation as well

as to speed up and improve decision making.

This chapter details some of the techniques

used to integrate data sets and to visualize this

integration. Two types of computer packages

have evolved to handle exploration and development

data: (i) Geographical Informations

Systems (GIS) for early stage exploration data,

usually generic software developed for other

nongeologic applications, discussed in section

DATA INTEGRATION AND GEOGRAPHICAL

INFORMATION SYSTEMS

to enable mine planning and resource calculations,

discussed in section .

INTEGRATION WITH RESOURCE

CALCULATION AND MINE PLANNING

SOFTWARE

What must be emphasized is that the quality

of data is all important. The old adage “rubbish

in and rubbish out” unfortunately still applies.

It is essential that all data should be carefully

checked before interpretation, and the best

times to do this are during entry of the data into

the database and when the data are collected.

A clear record should also be maintained of

the origin of the data and when and who edited

the data. These data about data are known as

metadata., and (ii) mining-specific packages designed