Page 266 - The Geological Interpretation of Well Logs

- THE GEOLOGICAL INTERPRETATION OF WELL LOGS -

coal correlates Y Zz
i MC

cys
A

shale cycle
sand bodies
DO NOT correlate correlates
Figure 15.20 Correjation of log shapes in a deltaic complex. From log shapes, the sand-body appears to be continuous but the
shapes only indicate similar facies. Incompatible fluids show that two separated sand-bodies exist. Persistent correlations are
based on coals and fine-grained facies. MC = marker coal, M = marker, CYC = marker cycle. W to Z is 3.5km.

Palaeontologicatly controlled correlations pod-like horst (the Buchan Horst) in the central North Sea
Palaeontological control in correlation is essential in (Burnhil) and Ramsay, 1981). The lithostratigraphy shows
sequence stratigraphy but was not sufficiently stressed some continuity but also inexplicable discontinuities. The
above (Section 15.2). This was done” delibérately to datings, fine enough for correlation, add sufficient infor-
emphasise the contribution of the logs. But any log study mation to explain the discontinuities and to corroborate
must be integrated with ‘event stratigraphy’ for reliable the continuous log correlations. The combined log and
correlation. That is, correlations are based on various bio- palaeontological correlations show a mid-Cretaceous,
logical events such as blooms or extinctions as well as the early Turonian unconformity over the horst itself with a
traditional appearances and disappearances of species later Turonian onlap. By correlating the unconformity
(e.g. Mitchener ef ai., 1992). Some events are considered with a regional Turonian regressive phase which occurred
to be isochronous (as far as the individual basin is throughout north-west Europe and North America, the
concermed) and therefore correlatable, others are facies authors show that the erosion was not a result of local
controlled. Because of sampling difficulties in the movement on the horst, but the result of a global change
subsurface, biostratigraphic events may not always be in sea level (Burnhill and Ramsay, 1981). The succeeding
identified in a well or be precise. The combined use of onlap had a similar cause. Clearly, this sort of reasoning
event stratigraphy and well logs is therefore essential. is not possible with undated log correlations.
Every specialist considers that his ‘discipline’ is the
most reliable and its results unassailable. The man who Correlations at outcrop
has spent his lifetime examining dinoflagellates, will Before a field is developed, the widely spaced exploration
argue their value to the denigration of any other informa- wells are correlated in detail over the reservoir interval and
tion. The man who has spent his lifetime analysing well predictions made for the infill producer wells about to be
logs will do the same for the logs! Of course, the essen- drilled. As all geologists know, there are always surprises
tial is to combine information using the strong points of during the infill drilling. Some intervals show unattended
each discipline. Frequently, specialists are unaware of (or continuity, others have unexpected variations. Work at
unwilling to admit to themselves) the weaknesses in their outcrop tends (o put a rather sober note on the possibilities
own speciality. for subsurface correlation, at Jeast in some facies.
This is simply to argue that much micro-biological The illustration chosen is of one very evocative piece
information has limitations in accuracy where correlation of work in which gamma ray profiles were logged at
is concerned. Well logs are, for the greater part, lithos- intervals along a quarry face by a gamma ray tool and a
tratigraphic records. When the two are combined, there logging truck! (Slatt et ai., 1992). The correlations sug-
is often a tendency to say that a particular horizon is gested by the log profiles can be compared with the actual
diachronous because microfauna occur at different levels. correlations seen and drawn from the outcrop (Figure
This is always a possibility, but distance between wells 15.22). The confidence which would justifiably exist after
must be considered. Wells a few kilometres apart are correlating the logs alone is misplaced, and perhaps
unlikely to have the same facies with different ages. A explains why development infill wells produce surprises.
careful balance in using the data must be achieved. The logs tend to simplify lithological responses, which in
The example illustrated shows such a balance between some instances is an advantage, in others it leads to
log correlation and biostratigraphy, nicely achieved errors. There is little that can be done about this except to
(Figure. 15.21). Middle Cretaceous rocks cover a small be aware of it.

256