Radiometric method - Geophysics for the Mineral Exploration Geoscientist - page 230

4.7.5 Interpretation of -logs

combined with differences in radioelement content within
the succession, makes for an exceptional example of litho-
logical/stratigraphic mapping using radiometrics.
Figure 4.26a shows a ternary image of the radiometric
data. The near layer-cake succession creates a pattern of
sub-parallel zones in the data which are associated with the
different outcropping lithotypes. Resolution of the stratig-
raphy is so good that it is possible to map the stratigraphic
succession accurately and identify lateral changes directly
from the radiometric data. The successions in four loca-
tions are shown in Fig. 4.26b .
The following features are indicated on Fig. 4.26a .
Where diapirs intrude the succession, the resulting cross-
cutting relationships are clearly seen (A). These units are
potassium-rich and therefore have a distinctive red appear-
ance. The layer-cake stratigraphy combined with radiomet-
ric responses which emanate only from surface materials
produce

γ

Logging variations in natural radioactivity on drillcore or
continuously downhole is done primarily to detect differ-
ences in lithology. Almost every type of geological environ-
ment is amenable to -logging. This is demonstrated by the

disparate geological settings from which the downhole logs
shown in Fig. 4.28 are taken, and the generally strong
correlation between the logs and the lithological variations.
In sedimentary rocks, it is usual for the

γ

γ -log to oscillate

between two , the higher level known as the shale

baseline, the lower known as the sand baseline ( Fig. 4.29 ). In
this respect the

' baselines '

-log is similar to the self-potential log (see
Section 5.5.3.2 ) . The shale baseline represents responses
primarily from K-bearing minerals such as mica. The sand
baseline re ects the subdued responses from clastic and
carbonate sediments which contain few radioelements. Both
the SP and

γ

-logs are useful indicators of facies/sedimentary
environments, which are inferred from changes in the
shapes of the log responses. For example, a fining-upwards
sequence is represented by a gradual change from the level
of the sand baseline to that of the shale baseline. Facies
analysis has been successfully applied in exploration for

γ

outcrop patterns of folding and faulting.
Particularly clear fold closures are labelled (B). Selected
faults are labelled (C); the lateral offset of the stratigraphy
is the main evidence for their presence. In several places
there are subdued radiometric responses from the faults
themselves, possibly related to sediments deposited by water
courses following the fault plane. Repetition of the strati-
graphic succession at (D) is indicative of thrust faulting.
In the east of the study area bedrock is concealed by
various types of unconsolidated cover. The radiometric
response of the cover is very different from the bedrock.
The pattern of responses reflects the local drainage, with a
creek bed (E) and various fans (F) evident. In some places
the bedrock which sourced the cover can be identified from
the similarity in the colour (radioelement content) of their
responses. The unconformity between basement and cover
can be traced (G), and cross-cutting relationships are evi-
dent in places.
The value of terrain models in the interpretation of
radiometric data is demonstrated by the images shown in
Fig. 4.27 . In Fig. 4.27a responses due to sediments in a
creek bed and adjacent slope deposits are well de ned; the
nearby exposed bedrock is clearly the source of detritus.
The relationships between topography and the individual
units comprising the bedrock are also seen in the radio-
metric data ( Fig. 4.27b ) . The good correlation between the
two data types is the main reason why the radiometric data
are able to map the geology of the rugged terrain accurately
over a large area, and it allows topographical features to be
related to lithotypes. The slope deposits in the east are
again clearly seen.

' model '

SP
(mV)

Natural gamma
(API)

10 50

- + - +

Baselines: Sand Shale Sand Shale

1 0 m

Sandstone
Mudstone

Limestone
Coal

Figure 4.29 Self-potential (SP) and

-logs through a coal-measure
sequence. Note the similarity of the two logs; both vary between sand
and shale baselines. Based on diagrams in Emery and Myers ( 1996 ).

γ