Ratcliffe et al., 2006 Chemostratigraphy of the TAGI Formation in

A REGIONAL CHEMOSTRATIGRAPHICALLY-DEFINED CORRELATION
FRAMEWORK FOR THE LATE TRIASSIC TAG-I FORMATION IN BLOCKS 402
AND 405A, ALGERIA
(This work has been submitted to The Geological Society of Londan and EAGE for possible publication.
Copywrite may be transferred without notice, after which this version may no longer be accessable.)
Ratcliffe, K.T., Martin, J and Pearce, T.J., Chemostrat Ltd, Unit 4, Llanfyllin Enterprise Park,
Llanfyllin, Powys, SY22 5DD, UK.
Hughes, A. D., Burlington Resources, Canary Wharf, One Canada Square, London, E14 5AA, UK.
Lawton, D.E., BHP Billiton Petroleum Ltd., Neathouse Place, London, SW1V 1LH, UK.
Wray, D. S. University of Greenwich, Dept. of Earth Sciences, Chatham Maritime, Kent, ME4 4TB, UK.
Bessa, F., Sonatrach, Division Petroleum Engineering et Development, 8 Chemin du Réservoir, Hydra, Algeria.
ABSTRACT: The Triassic Argilo-Gréseux Inférieur Formation (TAG-I) is one of the principal
hydrocarbon reservoirs in the Berkine Basin of Algeria. Sedimentological studies have shown that
it exhibits marked spatial and temporal facies variations on both a local field scale and regional
basinal scale. This variability, combined with a lack of diagnostic flora and fauna, make regional
correlation of the unit difficult. In turn, the lack of a consistent regional stratigraphic framework
hampers the comparison of the various correlation schemes devised by operators in the basin.
Contrasting the TAG-I in blocks 402 and 405a exemplifies the problems encountered when
attempting regionally to define a correlation framework for the interval. Between these two
blocks, a distance of approximately 200km, there are marked changes in the style of deposition
from sand-dominated, proximal fluvial systems in the SW (Block 405a, MLN, MLC and MLW
fields) to a more distal, more clay prone system in the NE (Block 402, ROD/BRSE/BSFN, SFNE
and BSF fields). A chemostratigraphic study of the TAG-I in these two blocks has allowed a fourfold correlation framework to be defined. Each chemostratigraphic package is geochemically
distinctive, with the geochemical variations being due to changes in detrital mineralogy of the
sandstones and changes in clay mineralogy of the claystones. Each of the chemostratigraphic
packages is bounded by a geochemically recognisable and regionally significant stratigraphic
surface. By combining the geochemical differentiation of the units and recognition of their stratal
boundaries, it is possible to define a correlation for the TAG-I between blocks 402 and 405a.
The proposed correlation between the two blocks suggests that, the northern parts of Block
405a may have been occupied by a spur or subsidiary channel from the main SW to NE trending
fluvial system, resulting in one of the chemically defined packages being demonstrably absent in
the MLNW, MLN, KMD and MLC fields when compared to the other areas of the study.
KEY WORDS: Chemostratigraphy, Lithostratigraphy, TAG-I, Fluvial, Sedimentology, Fluviolacustrine
ACKNOWLEDGEMENTS
The authors would like to express their thanks to
Joint Venture partners in the blocks 402 and 405a
areas, for allowing the release of well information
for publication, these are Sonatrach, ENI, BHP
Billiton, Burlington Resources and Talisman
Energy. We also wish to acknowledge that the
views of the authors may not fully reflect those of
the Joint Venture. The authors also thank Dr. Peter
Turner, Birmingham University, and Dr. David
Pilling, of PM Geos Ltd., for their comments made
on early versions of the manuscript.
INTRODUCTION
Triassic sandstones of the Berkine Basin in Algeria
(which include the late Triassic Argilo-Gréseux
Inférieur Formation, or TAG-I) are a prolific
hydrocarbon reservoir that is part of a SW to NE
trending fluvio-lacustrine depositional system,
extending eastwards from Algeria through southern
Tunisia (Figure 1). A summary of the regional
setting of Triassic sequences in the Berkine Basin
is presented by Busson (1971a, b) and an overview
of Palaeozoic petroleum systems by Boote et al
(1998). More detailed lithostratigraphy and
sedimentology of the TAG-I Formation are given
by Turner et al. (2001).
In blocks 405a and 402, the TAG-I thickness
varies from 25m to 100m, where it displays lateral
and vertical facies variations on a local and
regional scale. In Block 405a (which includes the
MLNW, MLN, KMD and MLC fields), the TAG-I
has a high net to gross ratio and is dominated by
sheet flood and braided, low-sinuosity fluvial
deposits, with laterally persistent sandstones and
claystones. In Block 402 (which includes the
ROD/BRSE/BSFN, SFNE and BSF fields), the net
to gross ratio is markedly lower, typically with low
Chemostratigraphy of the TAG-I Formation, Algeria
sinuosity fluvial deposits and occasional higher
sinuosity fluvial channel sandstones and relatively
working regional stratigraphic framework for the
TAG-I.
Figure 1: Location map and stratigraphic summary. The inset in the bottom right shows the approximate
extent of the Berkine Basin in Algeria and Tunisia and also indicates the area covered by the larger map.
On the main map, the oil fields that form part of this study are marked in green and the wells used to
define the regional correlation are marked. Gamma traces for the study intervals in selected wells are
displayed to demonstrate the change from sand-prone to more clay-prone facies in Blocks 405a and 402
respectively A stratigraphic summary in inserted in the upper right of the main map.
thick claystone intervals (Figure 1). In each block,
the respective joint ventures have erected detailed
stratigraphic zonation schemes that are valid within
their field areas (figures 2 and 3). However, a lack
of diagnostic fauna or flora from the well-bore
samples and the lateral changes in facies between
the two study areas, makes relating the
stratigraphic schemes proposed for blocks 405a and
402 problematic. Seismic resolution is also
insufficient to aid with regional correlations. This
paper demonstrates how the technique of
chemostratigraphy enables the TAG-I in blocks
405a and 402 to be correlated and thereby defines a
Chemostratigraphy, or chemical stratigraphy,
involves the characterisation and correlation of
strata using major and trace element geochemistry.
For this study, data for a total of 47 elements (10
major elements (Al2O3, SiO2, TiO2, Fe2O3, MnO,
CaO, MgO, K2O, Na2O and P2O5), 23 trace
elements (Ba, Be, Co, Cr, Cs, Cu, Ga, Hf, Mo, Ni,
Nb, P, Rb, Sc, Sn, Sr, Ta, Th, U, V, Y, Zn and Zr)
and 14 rare earth elements (La, Ce, Nd, Pr, Sm, Eu,
Gd, Tb, Tm, Dy, Ho, Er, Yb and Lu)) have been
determined using inductively coupled plasma
optical emission spectrometry and inductively
coupled plasma mass spectrometry. The sample
Chemostratigraphy of the TAG-I Formation, Algeria
Figure 2. Key geochemical features of well MLN-1, Block 405a, presented as geochemical profiles constructed from sandstone and
claystone data. MLN-1 displays all the features that are typical of wells from the MLN, MLW, KMD and MLC Fields. The
stratigraphic nomenclature for the MLN Field is displayed on the left of the wireline log data.
Figure 3. Key geochemical features of well BSFN-1, Block 402, presented as geochemical profiles constructed from sandstone and
claystone data. BSFN-1 displays many of the features that are typical of wells from the Block 402. The stratigraphic nomenclature for
the ROD area wells is displayed on the left of the wireline log data. The stratigraphic units defined by Turner et al (2001) for this well
are displayed on the right, together with the main geological features that these authors define at the Sequence boundaries. HERC =
Hercynian Unconformity, Discon = disconformity, Min Break = mineralogical break.
preparation and analytical procedures used in this
study are the same as those detailed in Jarvis and
Jarvis (1995), Pearce et al. (1999a) and
summarized in Pearce et al. (2005b). The study
incorporates data from over 1200 conventional core
samples from 35 wells in blocks 402 and 405a
Chemostratigraphy of the TAG-I Formation, Algeria
(Figure 1). Chemostratigraphy, as with most
stratigraphic techniques, is a subjective and
interpretative technique. In any rock section, even
within a single stratigraphic unit, the large number
of variables that potentially affect the elemental
concentrations means that not all samples will fall
within the “typical” geochemical signature defined
for that unit. This is exacerbated when samples
from core, due to inhomogeneities occurring on all
scales. In cuttings samples a certain degree of
homogenization occurs due to the nature of drilling
and cuttings mixing on their retrieval. Therefore, in
any chemostratigraphic study, a sufficiently large
dataset, such as that acquired for this paper, is
required to negate any “flyers” in the data.
Additionally, samples from a unit that plot outside
the typical field for that stratigraphic interval on
graphical plots may need to be disregarded for the
characterisation.
Sequences throughout the stratigraphic column
and from many of the world’s hydrocarbon
provinces
have
been
analysed
using
chemostratigraphy (Ehrenberg & Siring 1992,
Racey et al. 1995, Preston et al. 1998, Pearce et al.
1999a, 1999b, 2004, Wray 1999, Craigie et al.
2001, Pearce et al. 2003, Ratcliffe et al. 2004,
Pearce et al. 2005a, Pearce et al. 2005b). Although
the technique is used widely in the oil industry, the
majority of studies are proprietary and results
remain unpublished. This paper represents one of
the first attempts to use the technique as a
correlation tool to define major chemostratigraphic
packages that can be recognized over a distance of
200km in a regional continental setting.
STUDY SEQUENCE
In the study area, the TAG-I rests unconformably
on a subcrop of Carboniferous to Devonian
sediments and is overlain by the Triassic ArgiloCarbonaté (or Triassic Carbonates) (Figure 1). The
sedimentology of the TAG-I has been described by
Turner et al. (2001) to be a series of predominantly
fluvio-lacustrine facies that is overlain by the
estuarine to shallow marine Triassic Carbonates.
Turner et al. (2001) describe a total of 23
lithofacies based on core descriptions, with the
principal facies associations in the Block 402 area
consisting of fluvial channel sandstones, floodplain
siltstones and palaeosols, crevasse splay deposits,
lacustrine sediments and shallow marine
transgressive deposits.
Regionally, the internal lithostratigraphy of the
TAG-I is complicated, although locally it can be
layer cake in nature. While the apparent complexity
is, in part, inherent in a fluvial system with a wide
variety of facies associations, it is exaggerated in
the case of the TAG-I due to the lack of a regional
stratigraphic template into which more localised
schemes can be placed. Licences and field areas
shown in figure 1 are operated by a number of
different oil companies each having its preferred
stratigraphic framework that has been internally
devised. This makes stratigraphic comparisons
between blocks difficult. As demonstrated in this
paper, chemostratigraphy can be used to define the
principal stratigraphic building blocks of the TAGI on a regional scale.
Turner et al. (2001) presented a semi-regional
correlation of the TAG-I based on core
descriptions, mineralogical data and wireline log
responses. Their published study, however,
concentrated on the area associated with blocks 401
and 402 and did not extend into Block 405a, where
the TAG-I facies distributions are notably different.
They did demonstrate that the TAG-I thickens and
becomes more sandy to the SW. The
chemostratigraphic
correlation
framework
proposed here compliments and in most cases
confirms the stratigraphy presented by Turner et al
(2001), extending it into Block 405a and thereby
defining a stratigraphic framework that can be
applied in both the sand prone proximal and more
clay-prone distal settings.
CHEMOSTRATIGRAPHIC ZONATION
Four chemostratigraphic units, namely (in
ascending order) CP1, CP2, CP3 and CP4, are
defined for the TAG-I from blocks 402 and 405a.
Each of the units has internally consistent
geochemical features that allow them to be
differentiated from one another. Additionally, each
unit is bounded by regionally identifiable surfaces,
thereby allowing correlation between the two
blocks in question. The main geochemical features
of blocks 405a and 402 are shown on figures 2 and
3 respectively. Figure 2 uses the well MLN-1 as a
chemostratigraphic template for wells in Blocks
405a and figure 3 uses the well BSFN-1 for wells
in Block 402. The main mineralogical features that
are interpreted to affect the geochemistry in the two
areas are displayed on figures 4 and 5.
ZONE CP1
CP1 is the oldest TAG-I interval identified and is
separated from the underlying Palaeozoic
sediments by the Hercynian Unconformity. In
Block 402, CP1 is composed of thinly bedded
sandstones and silty claystones (e.g. BSFN-1,
Figure 3), whereas in Block 405a, the unit
comprises a basal sandstone that can be 12m thick
overlain by a claystone over 10m thick (e.g. MLN1, Figure 2). Despite this lateral variation between
the two blocks, both the claystones and sandstones
of CP1 can be geochemically differentiated from
those in the overlying units. In Block 402, this unit
is equivalent to Sequence 1 of Turner et al. (2001)
who describe the top of their Sequence 1 to be a
Chemostratigraphy of the TAG-I Formation, Algeria
Figure 4. Summary of mineralogical data (heavy mineral abundances, clay fraction XRD data and detrital mineralogy determined
from petrographic analyses) acquired from the MLN field in Block 405a. The wireline log displayed is that of well MLN-1, but the
mineralogical data have been acquired from a several wells in the MLN field. The position of the samples in the MLN-1 well are
estimated from local correlation of MLN wells.
regional disconformity in block 402, with the
sequence consisting of an infill of the inherited
post-Hercynian Unconformity topography relief.
The sandstones of CP1 are generally
differentiated geochemically from those of CP3
and CP4 by their low K2O/Al2O3 and high Zr/Nb
ratios (figures 2 and 3). The notable exception to
this is the lower sandstones in several wells from
the MLN Field (e.g. MLN-1 on figure 2), where
K2O/Al2O3 values are similar to those higher in the
study. These lower sandstones contain dark grey
mudstone rip-up clasts that are responsible for
locally increasing the K2O/Al2O3. The sandstones
of CP1 are geochemically similar to those of CP2.
Petrographic data indicate that the sandstones of
CP1 are relatively kaolinitic probably due to K
feldspars having been degraded, which probably
contributes to the low K2O/Al2O3 values in the
sandstones of CP1. The relatively high Zr/Nb ratio
values in the sandstones are shown, by heavy
mineral analysis, to reflect high zircon/rutile ratios
when compared to the younger sandstones (figure
4).
The claystones of CP1 are geochemically
differentiated from those above by their low
K2O/Al2O3
values
and
high
Al2O3/(CaO+MgO+Na2O+K2O) values. XRD
analyses demonstrate that the low K2O/Al2O3
values are reflecting the high kaolinite contents of
CP1 claystones compared to those within the
overlying chemostratigraphic packages (figures 4
and 5).
The top of CP1 is coincident with the
disconformity surface described in Block 402 by
Turner et al. (2001). The geochemical features
below this surface are typical of a pedogenically
altered and highly weathered surface. The Al2O3 /
(CaO+MgO+Na2O+K2O) ratios in the claystones
immediately below the disconformity are markedly
high (c.6-8) when compared to the claystones 510m below, where the values are typically less than
3 (figures 4 and 5). This is typical of palaeosol
development as described by Retallack (1997) and
by Pearce et al. (2005a). Additionally, the
TiO2/K2O in the claystones immediately below, or
close to the disconformity surface are markedly
high. This ratio reflects high anatase content
(Figure 4) in the relatively K2O-depleted
pedogenically altered, kaolinitic claystones. The
geochemical features of this surface are readily
recognised in blocks 402 and 405a, where they are
used to define the disconformity surface and
thereby regionally constrain CP1.
ZONE CP2
In Block 402, CP2 is directly equivalent to
Sequence 2 defined by Turner et al. (2001). It
comprises mostly claystones with relatively thin
(<1m) sandstones that overly the pedogenically
Chemostratigraphy of the TAG-I Formation, Algeria
Figure 5. Summary of mineralogical data (whole rock XRD data, clay fraction XRD data and detrital mineralogy determined from
petrographic analyses) acquired from wells BSFN-1 and ROD 8 in Block 402 .The wireline log displayed is that of well BSFN-1.
The position of the samples from ROD 8 have been placed in their correct stratigraphic position based on the proposed
chemostratigraphic correlation framework.
altered claystones at the top of CP1. CP2 is shown
below not to be recognised in the MLNW, MLN,
KMD and MLC fields of Block 405a.
The sandstones of CP2 have K2O/Al2O3 values
that are marginally higher than those of CP1 and
markedly lower than those of the overlying
sandstones (Figure 3). Petrographically, the
sandstones of CP2 show no evidence for feldspar
dissolution and yet contain markedly less feldspar
than those of the overlying units. This indicates that
a major influx of feldspar occurred between
deposition of CP2 and CP3. Turner et al. (2001)
describe a mineralogical break at their Sequence 2 /
Sequence 3 boundary where there is marked influx
of feldspar that indicates a change in sediment
provenance to a more feldspathic-rich lithology.
This change in mineralogy is clearly geochemically
modelled using K2O/Al2O3 values. The Zr/Nb
values in CP2 sandstones of Block 402 are high
compared to those of CP3, but are similar to those
of CP1 in Block 405a. This suggests that the heavy
mineral composition of CP1 is similar to CP2, but
is different to that of CP3, further suggesting a
change in sediment provenance prior to the onset of
CP3 deposition.
The claystones of CP2 are differentiated from
those below by their higher and upwardly
increasing K2O/Al2O3 values, but lower and
upwardly decreasing Al2O3 / (CaO+MgO+K2O
+Na2O) values (figures 3 and 5). They are
differentiated from the overlying claystones by
their high Zr/Nb values, mimicking the changes
seen in the sandstones.
In the north of Block 405a, (MLNW, MLN,
KMD and MLC fields), sandstones that lie
immediately on top of the geochemically defined
disconformity surface contain abundant feldspar,
resulting in high K2O/Al2O3 values. The sandstones
also have low Zr/Nb values, making them
geochemically similar to CP3, which is described
below (i.e. these sandstones postdate the change in
sediment
provenance
inferred
from
the
geochemical data at the CP2 / CP3 boundary in
Block 402). Therefore, the chemostratigraphic data
indicate that CP2 is absent from wells in the
northern Block 405a fields. The absence of CP2
sandstones is demonstrated graphically in figure 6,
where ternary diagrams for sandstone data in
selected wells, show that no sandstones with CP2
geochemical compositions are present in wells
MLN-1 and MLC-1. The unit can also be similarly
Chemostratigraphy of the TAG-I Formation, Algeria
shown to be absent from all study wells in the
MLNW, MLN and MLC fields. In the south of
Block 405a, where the MLSE Field (Figure 1)
occurs, the geochemical data from this area
indicates the reappearance of the CP2 unit (Figure
6).
The relatively thick CP1 sequence in the
northern Block 405a fields and the lack of evidence
in core for extensive erosion at the top of the CP1
unit in these fields implies that absence of CP2 is
due to non-deposition, rather than erosion. This is
supported by the presence of a palaeosol horizon at
the top of unit CP1 in these fields, which may be
the time equivalent of CP2. The non-deposition
could be attributed to a lack of accommodation
space, which in turn suggests that the base level
rise that allowed deposition of CP2 in Block 402
(and the MLSE area of Block 405a) did not affect
the area of the MLNW, MLN, KMD and MLC
fields. However, Turner et al. (2001), postulate that
during deposition of their Sequences 2 and 3 (units
CP2 and CP3 in this paper), the main fluvial
fairway ran SW to NE, but that there were a series
of spurs trending to the ENE and E. These spurs
entered the main fluvial fairway tangentially to the
main sedimentary depositional axis. It is therefore
suggested here, that the area of the northern Block
405a fields occupied one such spur during
deposition of CP2 in the main fluvial fairway.
Switching of fluvial drainage to another channel
complex could have resulted in the non-deposition
recorded in this area of Block 405a.
K2O
CP3
MLSE-3
CP2
CP 1
Al2O3
Sr
CP3
MLC-1
CP 1
CP3
MLN-4
CP 1
ZONE CP3
CP3 in Block 402 is lithologically variable,
comprising a mixture of fluvial sandstones
interspersed with overbank and lacustrine
claystones. In Block 405a, however, it is composed
mostly of a 5-12m thick sandstone unit that is
capped by a fining upward sequence culminating in
black shales (figure 1).
The sandstones of CP3 have higher K2O/Al2O3
values than those of CP2, but lower values than
those of CP4 (Figure 3). Petrography shows that
this change in K2O/Al2O3 values is directly related
to the feldspar content of the sandstones, which
reaches a maximum in CP4 (figures 4 and 5).
The claystones of CP3 are differentiated from
those of CP2 and CP1 by their high K2O/Al2O3
values, which are shown to reflect their more illitic
nature (figures 4 and 5). The claystones of CP3 are,
however, geochemically similar to those of CP4.
Although claystones of CP3 and CP4 are difficult
to differentiate chemically, the contact between the
two units is clearly marked by claystones with high
MgO/Al2O3 values (Figure 3), which is shown from
core inspection to represent the development of a
dolcrete in Block 402. In Block 405a, the claystone
at the top of CP3 has high MgO/Al2O3 values
CP3
CP2
SFNE-2
CP 1
CP3
CP2
BSFN-1
CP 1
CP3
ROD-2
CP2
Figure 6. Ternary diagrams constructed from
sandstone data in selected wells. These diagrams
show
that
sandstones
with
geochemical
composition of CP2 are absent from wells MLN-1
Chemostratigraphy of the TAG-I Formation,
Algeria
and MLC-1. selected wells. These diagrams show
that sandstones with geochemical composition of
CP2 are absent from wells MLN-1 and MLC-1.
relative to others in the TAG-I (Figure 2), but not
as high as those of the dolcretes in Block 402. In
cores from Block 405a, dolcretes at this horizon are
not described. However, the claystone that has
elevated MgO/Al2O3 values in Block 405a is a very
dark grey to black shale and has high illite contents
(Figure 4). The high illite contents are weakly
reflected by the relatively high K2O/Al2O3 values
in this claystone (Figure 2), but the marked change
in clay mineralogy is masked by the high Kfeldspar contents in this part of the TAG-I. The
presence of the dolcrete and illitic black shale at the
CP3 / CP4 boundary implies a prolonged and
laterally pervasive period of lacustrine deposition,
which suggests that the boundary is coincident with
the Sequence 3a / Sequence 3b boundary as defined
by Turner et al. (2001) in blocks 401 and 402.
ZONE CP4
In most wells, CP4 is composed of a 4-8m thick
sandstone that fines upwards into a claystone.
However, several wells, notably BSFN-1 in Block
402 have only limited sandstone development in
this unit. In Block 405a, the fining upward
sandstone typical of unit CP4, is occasionally
overlain by a 2-8m thick sandstone with a blocky
gamma-ray log motif (e.g. MLN-1, figure 1). This
sandstone is sufficiently geochemical different to
those below that CP4 is further subdivided into
CP4a and CP4b (figure 2) in Block 405a.
The sandstones of CP4a have the highest
K2O/Al2O3 values in the study interval, which is
petrographically shown to reflect high potassic
feldspar contents (figures 4 and 5). The sandstones
of CP4b have lower K2O/Al2O3 values, but higher
Zr/Nb values when compared to those of CP4a
(figure 2). Respectively, these geochemical features
are due to lower K feldspar, but higher zircon
values of this sandstone. The sandstone associated
with units CP4a and CP4b are interpreted to be late
stage, meandering fluvial system whose main
provenance was older TAG-I sediments. Reworked
older TAG-I sediments in these sandstones,
explains the decreased amount of the relatively
unstable feldspar (lower K2O/Al2O3 values) and
higher concentrations of the more stable zircon
grains (higher Zr/Nb values). The occurrence of
meandering fluvial sandstones at the top of the
TAG-I in Block 405a may indicate a rise in base
level associated with the on set of transgression
from the northeast.
As discussed above, there are no clear and
systematic geochemical characteristics that allow
the claystones of this unit to be differentiated from
those of CP3. Further, because of the meandering
nature of the CP4a and CP4b sandstones, some
well penetrations in Block 405a have encountered
no CP4 sandstone deposition. Thus CP3 claystones
are overlain by relatively thick sequences of CP4
claystones making exact correlation problematic.
Top TAG-I
Chemostratigraphically, the top of CP4 unit
corresponds to the top of the TAG-I. In blocks 401
and 402, Turner et al. (2001) place the top TAG-I
(top Sequence 3) at a transgressive surface of
erosion that comprises a lag deposit containing
intraclasts and bone/fish fragments. This is often
capped by a thin waxy black shale that contains
bivalves. Geochemically, this lag can be identified
in Block 402 by its high P and Mg contents (e.g.
ROD-8) and is therefore taken as the
chemostratigraphic top of the TAG-I. However, in
Block 405a no such lag has been recorded from
cored intervals, making definition of the formation
top more problematic. This is further complicated
when deposition of meandering fluvial sandstones
of CP4 does not occur, resulting in the TAG-I /
Triassic Carbonates formation boundary being a
claystone on claystone contact. Geochemically, in
Block 402, a local Rb/K2O minimum is
ubiquitously recorded at the transgressive lag
surface that marks the top of the TAG-I. A similar
Rb/K2O minimum is recorded at or close to the
topmost sandstone in wells from Block 405a. In
this block, this R/K2O feature is commonly
coincident with an upward decrease in Zr x Hf /
Al2O3 values. The Zr x Hf / Al2O3 ratio is a
sensitive indicator of the amount of silt grade
zircon grains present in the claystone and its
decrease at the top of CP4 reflects a decrease in
terrestrial sediment supply, such as may be
expected in association with a transgressive event
(i.e. the transgressive lag in Block 402). Therefore,
the Rb/ K2O minimum and upward decrease in Zr x
Hf / Al2O3 values are correlated with the
transgressive lag in Block 402 and are therefore
taken to be the regional marker for the top of the
TAG-I.
CONCLUSION
Chemostratigraphy is used to define four units
(CP1-4), each of which has internally consistent
geochemical compositions. Geochemical variations
in the sandstones are related to changes in detrital
mineralogy and provenance, while those in the
claystones are interpreted to be due to a mixture of
changes in detrital composition and changes in the
degree and type of syn-depositional weathering.
The bounding surfaces of each chemostratigraphic
package are related to regionally definable events:
•
Base CP1 boundary (base TAG-I) is
the regional Hercynian Unconformity,
which has been associated with a
Chemostratigraphy of the TAG-I Formation, Algeria
•
•
•
•
number
of
tectonic
events
(Hercynian/Variscan orogenies) that
have affected the Sahara craton
(Boote et al 1998).
CP1 / CP2 boundary is a regional
disconformity at which marked
pedogenesis has occurred, indicating
a period of non-deposition within the
TAG-I. It appears to have affected
large areas of the Berkine Basin and
is suggested here, to be a significant
break in sediment transport into the
study area.
CP2 / CP3 boundary is a major
change in sediment mineralogy that is
probably provenance related, due to
un-roofing, possibly, of an acidic
igneous complex of the hinterland.
This again implies either localised
tectonic activity in the source/
provenance area or major changes in
the drainage patterns feeding the axial
braided river system.
CP3 / CP4 boundary is a period of
regional lacustrine deposition (i.e. a
decrease in sand input) probably
caused by a reduction in basin profile
and/or stream bypass (to the southeast
of the study area). However,
punctuation
by
occasional
rejuvenation events is observed, with
sand being deposited in meandering
fluvial stream channels.
Top CP4 boundary (top TAG-I) is a
basin-wide transgressive event from
the northeast, with evidence of
increased marine influence in that
direction.
By enabling differentiation of each unit and
identifying
their
bounding
surfaces,
chemostratigraphy allows wells in blocks 402 and
405a to be correlated (figure 7). The proposed
correlation suggests:
•
The total TAG-I thickness is greater
towards the NE than in the SW, but
localised thickness variations can
occur between relatively closely
spaced areas such as the MLN and
MLC fields (figure 7). This in part is
due to localised thickening when
crossing a major bounding fault that
displays strike slip as well as vertical
movements. This fault complex forms
the main bounding fault not only for
Block 405a fields but also migrates
north-eastwards towards the giant
Ourhoud Field of Block 404 and 406.
•
•
•
•
CP1 is generally thicker in the MLN
and MLC fields compared to the
southeastern MLSE Field of Block
405a and the ROD fields within
Block 402.
CP2 is absent from the MLN and
MLC fields, which results in the
hiatal disconformity surface that
marks the top of CP1, being
coincident with the mineralogical
break that marks the CP2 / CP3
contact in the MLSE Field and Block
402 area.
The distribution of CP2 suggests that
the central Block 405a area (MLNW,
MLN, KMD and MLC fields) may
have occupied a different valley
system to that in which the TAG-I of
the MLSE Field and Block 402 fields
were deposited.
CP3 and CP4 are present in both
blocks and while CP3 appears to be of
relatively uniform thickness, CP4
appears to thin eastwards and
southwards from the central parts of
the MLN Field.
To conclude, it has been shown here that
chemostratigraphy can provide a working
framework for the comparison of lithostratigraphic
schemes devised for different TAG-I sections in
different parts of the Berkine Basin by different
field operators. The Basal TAG-I in Block 402 has
been proved to be equivalent to the Lower TAG-I
in Block 405a. CP2 is the Lower TAG-I in Block
402 and from the geochemical fingerprints of
different wells over the study area, is clearly absent
from some field areas in Block 405a. The Middle
and Upper TAG-I of Blocks 402 and 405a are
equivalent to one another and are assigned to CP3
and CP4 respectively. It is therefore important not
only to analyse sandstone but also claystone
samples, since both are required to build up the
correlative framework. This is particularly
important where sandstone deposition is replaced
by claystone sequences related to floodplain
sedimentation or periods of non-deposition and
pedogenic development.
Chemostratigraphy of the TAG-I Formation, Algeria
32 90
Carbonate
28 00
32 70
CP4
28 10
CP4
CP4
CP3
CP4
32 80
32 90
34 00
32 40
CP3
CP4
32 60
34 10
CP2
29 00
28 60
33 10
CP1
CP4
28 70
CP4
CP4
CP3
CP3
29 90
28 80
CP3
CP3
Chemo
Regional
Chemo
Regional
CP3
29 10
CP3
CP2
33 00
29 00
CP3
29 10
29 00
29 80
30 10
29 00
CP2
CP2
CP1
33 50
CP1
32 90
34 40
CP1
33 10
33 20
28 70
33 00
29 20
CP2
29 20
30 30
29 40
CP1
CP1
CP1
33 70
28 80
Carb
on
CP1
Carboniferous
29 50
29 60
29 60
29 40
30 20
29 40
CP1
30 50
33 20
29 50
D-2
RO
D-1
RO
-1
-1
-2
FE
FN
SN
BS
N
ML
W-1
ML
29 50
SE
-2
C
ML
D-8
RO
BR
N-4
ML
-2
5a
40
CP1
29 50
33 10
ROD
BSF
29 40
CP1
29 30
30 40
29 30
33 30
iferou
s
33 40
-3
-1
SE
ML
SE
ML
Approx Field outlines
29 30
29 20
CP2
30 00
30 10
SFNE
CP2
29 30
CP1
CP1
33 00
KMD
29 20
30 20
29 10
32 90
33 60
MLWN MLN
CP2
29 10
32 80
29 10
CP2
CP3
CP4
N
Study wells
30
29 00
30 00
28 90
32 90
29 90
Km
CP4
28 90
28 90
29 70
32 80
34 30
MLC
28 80
CP4
28 70
29 80
CP3
CP3
32 70
33 20
0
Depth
32 70
CP1
CP1
MLSE
Gamma API
28 70
28 90
32 80
CP3
CP3
33 30
CP1
Depth
28 80
CP4
29 50
32 60
33 40
CP1
Gamma API
28 80
28 90
28 50
Depth
28 70
29 60
33 00
34 20
Gamma API
29 70
28 60
32 50
CP3
28 40
Depth
29 60
32 70
33 20
CP2
Gamma API
32 50
CP3
28 30
CP2
Depth
ROD-2
28 60
CP4
CP4
32 40
CP4
CP4
28 70
28 80
Gamma API
28 50
Carbonate
32 50
32 60
33 10
CP3
Depth
29 40
CP3
28 20
Gamma API
32 30
33 00
33 90
28 60
Depth
32 30
33 80
CP4
Gamma API
29 30
28 40
CP4
Depth
ROD-1
32 20
33 70
28 50
Gamma API
BSFN-1
Chemo
Regional
Depth
BRSE-2
Chemo
Regional
Gamma API
ROD-8
Chemo
Regional
Depth
SFNE-2
Chemo
Regional
Gamma API
MLC-2
Chemo
Regional
Depth
MLC-1
Chemo
Regional
Gamma API
MLN-3
Chemo
Regional
Depth
MLN-1
Chemo
Regional
Gamma API
MLN-4
Chemo
Regional
Depth
MLW-1
Chemo
Regional
Gamma API
MLSE-3
Chemo
Regional
Depth
Chemo
Regional
MLSE-1
Line of section
Figure 7. Correlation of selected wells in Block 402 and 405a. The wells selected for this diagram typify the wells of Blocks 402 and 405a.
Chemostratigraphy of the TAG-I Formation, Algeria
CP1
Gamma API
REFERENCES
Boote, D.R.D., D.D. Clark-Lowes and M.W. Traut,
1998. Palaeozoic petroleum systems of
North Africa. In D.S. Macgregor, R.T.J.
Moody and D.D. Clark-Lowes (eds),
Petroleum Geology of North Africa,
Geological Society Special Publication
No132, p. 7-69.
Busson, G. 1971a. Le Mesozoique sarahien,
deuxieme partie: Essai de syntheses des
donnees de sondages Algero-tuniiens,
Centre de recherche sur le zones arides,
serie geologie no. 11, Editions de Centre
NationAl2O3de la recherche scientific, 340
pp
Busson, G. 1971b. Principes, methodes et resultats
d’une etude stratigraphique du Mesozoique
saharien. These, Universite Pierre et Marie
Curie, Paris, France 441p
Craigie, N. W., A. Hughes, J. Valentine, and M.
Dix,
2001.
Applications
of
chemostratigraphy (abs): Petroleum Society
of Great Britain Newsletter Aug/Sept 2001,
p. 9-11.
Ehrenberg, S.N and E. Siring. 1992. Use of bulk
chemical
analysis
in
stratigraphic
correlation of sandstones: an example from
the Statfjord field, Norwegian continental
shelf. Journal of Sedimentary Petrology, v.
62, p. 318-30
Jarvis, I., and K. E. Jarvis, 1995, Plasma
spectrometry in earth sciences: techniques,
applications and future trends, in I. Jarvis,
and K. E. Jarvis, eds., Plasma Spectrometry
in Earth Sciences: Chemical Geology, v.
95, p. 1-33.
Martin, J., Hughes, A. D, Lawton, D.E., Pearce, T.J.,
Ratcliffe, K.T. 2004. The Triassic Argilo
Gréseux-Inférieur (TAG-I) of Algeria:
high-resolution chemostratigraphy as a
correlation tool in proximal to distal fluvial
facies. (abs). AAPG Conference, Dallas
2004.
Pearce, T. J., B. M. Besly, D. S. Wray, and D. K.
Wright, 1999a. Chemostratigraphy: a
method to improve interwell correlation in
barren sequences - a case study using
onshore
Duckmantian/Stephanian
sequences
(West
Midlands,
U.K.):
Sedimentary Geology, v. 124, p. 197-220.
Pearce, T. J., J. Martin, and D. S. Wray, 1999b,
Chemostratigraphic characterisation of
Early Tertiary sequences: West of
Shetlands. The North Atlantic Igneous
Province:
magmatic
controls
on
sedimentation (abs), Geological Society of
London Meeting May 1999
Pearce, T. J., D. McLean, D. K. Wright, C.J. Jeans
and E.W. Mearns 2005a. Stratigraphy of
the Upper Carboniferous Schooner
Formation,
southern
North
Sea:
chemostratigraphy, mineralogy, palynology
and Sm-Nd isotope analysis. In J.D.
Collinson, D.J. Evans, D.W. Holliday and
N.S.
Jones
(eds.),
Carboniferous
hydrocarbon geology: the southern North
Sea and surrounding onshore areas,
Yorkshire Geological Society, Occasional
Publications series v. 7 p.165-182.
Pearce, T. J., D. S. Wray, K. Ratcliffe, D. K. Wright
and
A.
Moscariello,
2005b.
Chemostratigraphy
of
the
Upper
Carboniferous
Schooner
Formation,
southern North Sea. In: J.D. Collinson, D.J.
Evans, D.W. Holliday and N.S. Jones
(eds.), Carboniferous hydrocarbon geology:
the southern North Sea and surrounding
onshore areas, Yorkshire Geological
Society, Occasional Publications series v. 7
p.147-164.
Preston, J., A. Hartley, M. Hole, S. Buck, J. Bond,
M. Mange, and J. Still, 1998, Integrated
whole-rocK2O trace element geochemistry
and heavy minerAl2O3chemistry studies:
aids to correlation of continentAl2O3redbed reservoirs in the Beryl Field, UK2O
North Sea: Petroleum Geoscience, v. 4, p.
7-16.
Ratcliffe, K. T., A. D. Hughes, T. J. Pearce, and J.
Martin,
2002,
Enhanced
reservoir
characterisation of the Triassic Argilo
Gréseux-Inférieur, Algeria using high
resolution chemostratigraphy (abs). AAPG
Conference, Salt Lake City 2002.
Ratcliffe. K.T., A.M. Wright, C. Hallsworth, A.
Morton, B. A. Zaitlin, D. Potocki and D.S.
Wray, 2004. An example of alternative
correlation
techniques
in
a
low
accommodation
setting,
non-marine
hydrocarbon
system:
The
(Lower
Cretaceous) Mannville Basal Quartz
succession of S. Alberta. American
Association of Petroleum Geologists,
Bulletin, v. 88, p 1419 - 1432
Racey, A., M.A. Love, R.M. Bobolecki and J.N.
Walsh. 1995. The use of chemical element
analsyis in the study of biostratigraphically
barren sequences: an example from the
Triassic of the central North Sea (UKCS).
In, R.E. Dunnay and E.A. Hailwood (eds.),
Non-biostratigraphical methods of dating
and correlation, Special Publication no 89,
Geological Society, London p. 69-105.
Wray, D. S., 1999, Identification and long-range
correlation of bentonites in Turonian Coniacian (Upper Cretaceous) chalks of
northwest
Europe:
Chemostratigraphy of the TAG-I Formation, Algeria
GeologicAl2O3Magazine v. 136(4), p. 361371.
Retallack, G.J. 1997. A colour guide to paleosols.
John Wiley and sons Ltd., Chichester,
England175pp.
Turner, P., Pilling, D., Walker, D., Exton, J., Binnie,
J. and Sabaou, 2001. Sequence stratigraphy
and sedimentology of the late Triassic
TAG-I (Blocks 401/402, Berkine Basin,
Algeria), Marine and Petroleum Geology,
v. 18, p. 959-981.
Chemostratigraphy of the TAG-I Formation, Algeria