Page 276 - The Geological Interpretation of Well Logs

- THE GEOLOGICAL INTERPRETATION OF WELL LOGS -

(for a good description of time series analysis see 16.6 But is it geology?
Doveton, 1994). By using spectral analysis techniques on
Al that same SPWLA meeting in Paris mentioned at the
a log trace, an attempt is made to mathematically extract
beginning of this chapter, there was an old French lady,
the dominant scales of time dependent cyclicity, which
quite tall, sprightly and speaking perfect English. She
are generally in the Milankovitch frequency band of
remembered as a litile gir) going with her father to the
10,000-100,000 years (Motine and Ogg, 1990). The
basement of the Ecole des Mines in Paris to see some of
methods are especially suited, for the moment, to fine
his strange machines, all wires and coiis and tubes. As
grained detrital sediments deposited in environments
the younger daughter of Conrad Schlumberger, she could
where the sediment record is essentially complete and
never have imagined that her father would create within
sedimentation rates constant. The example shown is of an
her lifetime, with his brother Marcel and friend Henri
interval of Cretaceous (upper Albian) mari from a cored,
Doll, the company that is Schlumberger and the now
shallow borehole in northern Germany (Kirchrode 1/9])
very sophisticated technique of well logging. Predicting
with quite fast sedimentation rates (10cm/1000yrs) and
the future it seems, is of significantly less use than having
representing 2.5 million years (from biostratigraphic
confidence in the present.
dating, BCCP Group, 1994). The gamma ray log of the
The genius of the Schlumberger brothers and Henri
interval has been processed by wavelet-transform using
Doll was involved in a field of its own making. Now the
variable ‘windows’ (Prokoph and Thurow, 1995) and
field has many workers, many contributors. But that does
re-displayed as a wavelet-scalogram (Figure 16.5), The
not mean that it has developed beyond invention.
plot makes identifiable the thickness (wave-length) of
Charting the introduction of new logging tools year by
evident cycles which, if the sedimentation rate is known,
year shows an interesting picture. From the early days
can be given a time score. For example, the strong, 12m
in the 1920s to about the tate 1970s there was a progres-
thick period seen between 35m—163m can be equated
sively steepening climb with many new inventions. A
with a 100,000 year eccentricity cycle (based on sedi-
plateau occurred from the late 70s to the mid 80s with
mentation rates of LOcn/LO0Q0yrs). This strongly cyclic
little that was new. The oil crisis intervened. Then there
section is interpreted from biostratigraphic work to have
was another steep rise, so that at present, the rate of
been deposited during a sea level highstand (indicated by
introduction of new tools is bewildering.
the foraminiferal analyses) when global rather than local
Again that SPWLA meeting. In the exhibition halls,
influences were dominant (BCCP Group, 1994}. Clearly,
most of the large computer screens belonging to the
such a quantitative analysis of a log trace leads to far
service companies had log generated images on them.
greater and more reliable interpretation possibilities than
Geological images. But how many geologists were there
a simple, casual, qualitative inspection.
to see? Too few. And yet here is a huge bank of sub-
Despite this example, software which treats the purely
surface geological data, in detail never dreamed of by
geological aspects of logs is being ‘eft behind by the
the Schlumberger brothers: and from a range of sensors
considerable advances in software designed to improve
they could only have hoped for. But of course it all needs
tool response, signal processing and petrophysical
interpreting, and this demands computers, applicable
modelling and interpretation. What a geologist can do
software, new ideas and a brave mind. The geological
with his log data depends on the software that he has at
hammer is past: it isn’t necessary to hit things anymore to
his disposa]. At the moment this is limited. There is a lot
understand them. A geologist today is only as good as his
of progress to be made in the right direction, but the
software.
possibilities are enormous.
But it is still geology.

266