Geophysical data acquisition, processing and interpretation - Geophysics for the Mineral Exploration Geoscientist - page 31

variations to be properly represented (at least up to the
Nyquist frequency of the sampling interval), but the much
greater distance between lines, i.e. greater across-line sam-
pling interval, means similar variations in the across-line
direction are likely to be under-sampled (aliased). This is
often less of a problem that it might seem because orientat-
ing the survey lines perpendicular to strike ensures there
will be a greater degree of variation in the geophysical
response along-line than in the strike-parallel across-line
direction. Of course, for responses that are equidimen-
sional, the sampling interval should ideally be the same in
both directions in order to produce a properly sampled
anomaly. In reality, it is the (wider) line spacing that deter-
mines overall survey resolution.
For airborne surveys, the nature of the topography in
relation to the survey parameters will determine the survey
platform and the cost of the survey. Terrains of moderate
relief can be surveyed with small

discovery of the deposit. The line spacing is typical of a
modern reconnaissance aeromagnetic survey, although the
survey height is signi cantly greater than that of current
practice. The data have been gridded (see Section 2.7.2 )
and then imaged and contoured (see Section 2.8 ) . The
important aspects of Fig. 2.12 are that the observed
response of the mineralisation varies in amplitude and that
the locations of the maximum and minimum responses
change with line spacing and the location of the survey
lines. On using subsets of the data by line spacing and line
location, quite significant changes can be seen in the amp-
litude of the measured anomaly, the apparent location of
the source, and its strike length and direction. When the
line spacing is large compared with the lateral extent of the
actual anomaly ( Fig. 2.12a and b ), the peak responses are
located on the nearest line(s) and the amplitude of the
anomaly is underestimated. They are only correctly located
when the line fortuitously passes over the centre of the
actual anomaly ( Fig. 2.12c to h ) , and then measured ampli-
tudes more closely re ect the actual amplitudes. Every
aspect of the anomaly (i.e. its amplitude, shape, gradients
and trend direction) improves with optimally positioned
lines and decreasing line spacing. The worse-case situation
occurs when the anomaly lies entirely between two lines, in
which case no anomalous response will be detected.
The Marmora example demonstrates clearly the import-
ance of setting the line spacing to suit the across-line width
of the anomaly. It is useful during survey design to deter-
mine the probability of detecting a target of particular
dimensions with a particular survey configuration. The
probability of detecting an anomaly depends on the distri-
bution of the stations (or data points) and the orientation of
the survey lines, assuming that the line spacing and the
station interval are different, with respect to the dimensions
and shape of the target

fixed-wing aircraft, whilst
helicopters are appropriate for low-level surveying of
mountainous areas (Mudge, 1996 ) . Cost and logistical
considerations of course will ultimately determine the nal
survey parameters; and to that extent, the line spacing
should be increased first and, if necessary, followed by
limited increase in survey height.
Unmanned airborne survey platforms are likely to
become available for commercial use in the near future
(McBarnet, 2005 ). The advantages of using an unmanned
aircraft include reduction in cost, the ability to fly lower
and in poorer visibility than is possible with manned
aircraft, and the possibility of surveying at night when
environmental noise levels are lower.

2.6.4 Feature detection

When the aim of the survey is to identify anomalous
responses, the survey parameters should be set to maximise
the chance of these responses being measured, and prefer-
ably on several survey lines to con

s response. The area of detectable
response should not be confused with the surface projection
of the target itself because the geophysical response nor-
mally extends over an area larger than the target itself
( Fig. 2.4 ), which helps enormously in aiding detection.
Of course, the target

'

rm the validity of the

response.
Agocs ( 1955 ) shows how the measured anomaly varies
with station locations, using aeromagnetic data from the
Marmora Fe deposit, located in Ontario, Canada. The
deposit is a magnetite skarn which produces a very distinct
magnetic anomaly against the quiet background response
of the host carbonate sequence. The data shown in
Fig. 2.12 were acquired along parallel survey lines spaced
approximately ¼ mile (400 m) apart at a nominal terrain
clearance of 500 ft (152 m), and were responsible for the

s response can be complicated and
needs to be recognisable above the noise to ensure its
detection. This raises the issue of what constitutes detec-
tion. A single anomalous measurement could well be mis-
taken for noise, so at least two readings, but preferably
more, need to be made in the area of a detectable response.
Galybin et al.( 2007 ) investigate the problem of recog-
nising an arbitrarily oriented elliptical anomaly with a
survey of given line interval and station spacing. The

'