You are on page 1of 10

AGSO Journal of Australian Geology & Geoph ysics, 17(5/6), 67- 76 Commonwealth of Australia 1999

The problem of inconsistency between thermal maturity indicators used for


petroleum exploration in Australian basins
Ronald W. T. Wilkins I
A major frustration in thermal maturation modelling for petro technical, methodological and conceptual problems inherent in
leum exploration in Australian sedimentary basins is the inconsis each technique. When the differences between the concepts of
tency between the values of different thermal maturity indicators. rank and thermal maturity are considered, it can be shown that
Vitrinite reflectance (VR) , Rock-Eval T m" , spore colouration in some inconsistencies are more apparent than real. It is important
dex (SCI) and fluorescence alteration of multiple macerals (FAMM) to consider this distinction when selecting data against which to
for wells from three Australian basins show inconsistencies due to model burial and thermal histories.

Introduction
The driving force for modern thermal maturity (TM) studies is Keiraville Konsultants (KK)' and Robertson Research Interna
the demand for high- quality data to constrain thermal history tional (RRI). For many North West Shelf wells, depth vs VR
models. Because no known TM indicator is universally reliable profiles determined by these laboratories are the same within
(Whelan & Thompson-Rizer 1993), new approaches are being experimental error. For some wells, however, there are clear
continually devised to cope with the known problems of widely discrepancies in the VR results, which may ultimately have
used techniques. The normal progression of a new technique is originated in the different methods of sample preparation used,
through testing against an established TM indicator, usually that is according to whether the mounts were prepared as whole
vitrinite reflectance (VR), on sediments of different age, geo rock samples (KK) or as kerogen concentrates (RRI). For rocks
graphical distribution and depositional environment, as well as containing sparse organic matter, kerogen concentrates have the
contrasting structural settings and thermal regimes . Depending advantage of enabling statistically adequate numbers of vitrinite
on the apparent range of application determined in this demon grains to be more readily measured. However, this advantage is
stration period, the technique will become established among balanced by the fact that contaminant organic matter and pyrite
the pantheon ofTM indicators in common use or it will eventu are also concentrated by this preparation process. The main
ally become replaced by new thermal maturity tools. advantage of using whole rock samples is that the indigenous or
Burnham & Sweeney (1989) commented that a major im first generation vitrinite can be distinguished with much greater
pediment to improving their model of vitrinite maturation for confidence. Where discrepancies exist in VR measured by dif
global application was the inconsistency between various VR ferent laboratories it is likely that difficulty in the identification
data sets . The problem extends to relationships with values of the indigenous vitrinite population is the major source of
from other TM techniques. Even among results for well-estab error.
lished TM indicators, there may be a surprising degree of incon Apart from technical matters, there are some important theo
sistency. This is perhaps a particular problem for petroleum retical questions on the VR technique which are not, as yet,
exploration in Australia, because the region may have been out answered satisfactorily. It has not been proved that vitrinites
side the range of initial testing of a technique. Indeed, for Aus with the same reflectance from sediments of different age have
tralian wells, it is commonplace for TM indicators to be in the same rank. Ting & Sitler (1989), in a comparative study of
partial or even gross disagreement. Because there are few gener North American Carboniferous and Cretaceous coal, presented
ally accepted criteria for the assessment of inconsistent results, evidence that vitrinite in these coals does not have the same rank
the ultimate choice of data on which to base models may be at the same value of reflectance. On the other hand, it is well
quite subjective. In this paper an attempt is made to resolve the known that serial samples through coal seams commonly dis
inconsistencies in some results from three of the most com playa range ofVR, although they have the same thermal history
monly applied TM indicators (VR, Rock-Eval Tmax and spore and are, therefore, ofthe same rank (e.g. Diessel1992, Newman
colouration index- SCI) and the recently developed fluores et al. 1994, Quick 1994). Van Krevelen (1993), in a comparison
cence alteration of multiple macerals (FAMM) technique, using of chemical and VR data on coals from different continents and
some North West Shelf wells as examples. ranging in age from Carboniferous to Tertiary, concluded that
while VR is an excellent parameter of relative rank, it is neither
Relationships between TM parameters universal nor absolute.
A related question is the effect of inherited composition of
Vitrinite reflectance the vitrinite precursors from the time of diagenesis on the reflec
McCulloch & Naesser (1989), among others, have commented tance of telovitrinites, even of the same age. This gives rise to
on the unique place of vitrinite reflectance among petrographic the problem of vitrinite reflectance suppression for hydrogen
and geochemical techniques for the study of the thermal history rich vitrinite (perhydrous) compositions (Hao & Chen 1992).
of sedimentary rocks. Indeed, VR is widely regarded as the The main problem in assessing the importance of this phenom
primary TM indicator (Whelan & Thompson-Rizer 1993) against enon has been the lack of any easy method to quantify the effect
which all others are assessed. As kinetic models for VR (Burnham by direct measurements on individual vitrinite grains. It has long
& Sweeney 1989, Sweeney & Burnham 1990, Suzuki et al. been known (Stach et al. 1975) that high fluorescence intensity
1993) are now well established and widely tested with commer is a characteristic of perhydrous vitrinites, and that these
cial thermal maturation modelling programs, such as BasinMod, vitrinites have an anomalously low reflectance; but it is only
it is common for TM data from other indicators to be trans since the recent introduction of the FAMM technique (Wilkins
formed into equivalent YR. et al. 1992b, 1995) that quantitative assessment has been rou
VR measurements on Australian North West Shelf samples tinely possible. Although the importance of the problem was
have been carried out in a number of independent laboratories flagged by Price & Barker in 1985, largely on the basis ofreflec
both in Australia and overseas. Two laboratories that have made tance anomalies in certain data sets, it is only now becoming
outstanding contributions to North West Shelf TM studies are clear from FAMM studies (Ellacott et al. 1994, Wilkins et al.

, CSIRO Division of Petroleum Resources, POBox 136, North , As subcontractor to Analabs and , more recently, Geotech.
Ryde 2113 , Australia
68 R.W.T. WILKINS

I 992a, 1994b, 1995, and CSIRO unpublished reports) that the organic matter. This is because there is strong evidence (Wilkins
magnitude of the problem has been seriously underestimated, et al. I 994b) that vitrinite reflectance suppression is extremely
especially for Jurassic and Cretaceous source rocks. As a result common in the marine and especially the marginal marine Creta
of considerations such as these, McCullough & Naesser (1989) ceous and Jurassic rocks ofthe North West Shelf. In fact, in RRI
commented in their essay that VR, as the preeminent standard results on the Carnarvon Basin samples, there is a close rela
in normal use, seems to be on shaky ground. If this assessment tionship between VR and SCI (see section on Flamingo-I),
is correct, it would appear to pose a very serious problem in which could imply that the two parameters determined from
deed, potentially affecting all other techniques which have been kerogen concentrates are not independently obtained, but, rather,
calibrated against YR. that each result is influenced by the other. It may be argued that
the results from both techniques benefit from this internal com
Vitrinite reflectance and spore colouration index parison. In any case, it is important to bear this question of
There is no doubt of the value of the colour of palynomorphs as methodology in mind when VR results from different laborato
a TM indicator (Staplin 1977, 1982). Haseldonckx (1977) has ries are compared.
discussed the correlation of different palynomorph colour index Although some effort has been put into establishing appro
scales, of which the thermal alteration index (TAl) and SCI are priate kinetics for spore colouration (Peters et al. 1977), they
the most commonly used. Some of the disadvantages of the are not well enough known for SCI data to be used directly and
method are the subjectivity in identifying the indigenous confidently in thermal maturation modelling. Furthermore, the
palynomorph population in kerogen concentrates (although the problem of transforming SCI into equivalent VR is consider
possibility of dating the palynomorphs may assist in identify able. Cooper (1977) discussed the relationship between SCI
ing those that are reworked), variable exine thickness, bleaching and VR and showed that it is not universal, but depends at least
and staining effects, and the low resolution of the method in the upon the age of the organic matter and the thermal history of the
zone of petroleum generation (Jones & Edison 1978, Senftle et basin. Cooper's SCI to VR transformation curves for several
al. 1993 , Petersen & Hickey 1985). Because of these factors, groups of samples are shown in Figure 1 together with the RRI
and especially the last, it is difficult to ascertain by how much calibration curve for the North West Shelf, Western Australia
the method is affected by changes in organic facies. (Robertson Research Australia 1988). It should be remembered
Schwab et al.(1994 ) have argued that the TAl technique has that at the time these calibrations were determined, the impor
the considerable advantage over VR that, since it is based on the tance of the vitrinite reflectance suppression effect was not
hue and intensity of hydrogen-rich organic constituents in widely recognised and it is likely that the differences between
transmitted light, the results are not, like vitrinite reflectance, the calibrations are partly due to this factor.
affected by a suppression effect. He found that spore colour
Vitrinite reflectance and fluorescence alteration of
did not reflect important organic facies changes in sections
including Pennsylvanian black shales with elevated hydrogen multiple macerals
index (HI) values relative to the adjacent sediments. There is The fluorescence alteration of multiple macerals (FAMM) tech
also evidence from North West Shelf wells (see section on nique (Wilkins et al. 1992b, 1995) is a powerful new TM tech
Bowers-I) that the spore colour is not affected in major zones nique designed to cope with the phenomenon of suppression of
in which the vitrinite reflectance is suppressed. On the other vitrinite reflectance and the problems posed by organic matter
hand, if spore colour is insensitive to factors causing the sup in which the identification of vitrinite is equivocal. Because of
pression effect in vitrinite, a very scattered relationship be the suppression effect, the values of measured VR and equiva
tween VR and SCI would be expected for North West Shelf lent VR from FAMM (which is internally corrected for sup-
5 A EqVR% 5 B EqVR %

0.40 , -_ _ _ _---, ... 0.40
_ Vitrinite
o Other macerals


0.50 ............................. 0.50
a
:;:J
EqVR% =0.53
~
c
a
0.60
~L-

~ ..................................... 0 .70
l\l .
0.70
Q)
U Tertiary 'normal' Permian 'normal '
C vitrinite line - - vitrinite line --
Q)
O.M .............................. 0.80
U

-
rJ)
Q)
o
a
L-

0.90
:::l
u:::: Subhydrous vitrinite Perhydrous vitrinite
EqVR% = 0.94 <- - -_
..0 1.00 e . - ............................... 1.00
o
........................................................................................... 1.10 ..... ~ .. ................................................................. 1.10

.... .. .. ............................................................................. 1.20


0 ................... 1.20
0.5
10 2 10 3 10' 10 5
Fluorescence intensity at 400 s Fluorescence intensity at 400 s

Figure 1. Relationship between spore colouration index and vitrinite reflectance for rocks of different sample
groups: 1, Cretaceous-Tertiary; 2, Triassic-Oligocene; 3 , Carboniferous coals; 4, Late Tetiary with high initial
temperature gradient; 5, North West Shelf Australia (modified from Cooper 1977 and Robertson Research Australia
1988).
THERMAL MATURITY lNDICATORS 69

pression) commonly differ by as much as 50% for Jurassic and monly numerically similar, indicating compositions which are
Cretaceous rocks of the North West Shelf, and differences of close to orthohydrous, except where the environment has been
100% are known. marine influenced such as that represented by the Late Triassic
One of the most powerful features of the FAMM technique Brigadier Formation.
is the systematic relationship between VR and FAMM data.
This provides a test of consistency for VR and equivalent VR Vitrinite Reflectance and Tmax
values, which is illustrated in Figure 2. Individual macerals may
be plotted on this diagram using parameters derived from their Rock-Eval pyrolysis results are available for some North West
curves of fluorescence intensity against time (Wilkins et al. Shelf wells for which VR has not been determined. T max' the
1992b, 1995). The sub-vertical lines represent maturation path temperature at maximum rate of hydrocarbon generation during
ways for telovitrinites of different hydrogen-richness and mag artificial pyrolysis, is an important TM parameter, but its use
nitude ofVR suppression. Horizontal lines on the diagram are is critically dependent upon the correct identification of kero
iso-rank lines. The diagram indicates that VR and FAMM equiva gen type (Espitalie et al. 1985). In the Rock-Eval technique,
lent VR are numerically the same only if the composition of the kerogen typing is mainly accomplished through the hydrogen
vitrinite is 'normal' (orthohydrous). For all other vitrinite com index (HI) parameter, given by 100 S, / TOC in mg g-' TOC,
positions, perhydrous and subhydrous, a correction factor is where S, is the hydrocarbon generated by pyrolysis and TOC is
required to be respectively added to or subtracted from the the total organic carbon in wt%. Horstman (J 994) has pointed
measured YR. out that the oil potential of source rocks containing organic
For example, in Figure 2, a vitrinite plotting at ' B ' has a matter with a high inertinite content is systematically underes
measured reflectance of 0.50%, and a FAMM equivalent VR of timated by Rock-Eval pyrolysis . This is because inertinite con
0.70%. The suppression iso-correction curves indicate that a tributes little to the hydrocarbons released on pyrolysis, yet it
suppression correction of + 0.20% should be applied to the is measured in the TOe. An adjusted hydrogen index Hl. which
measured VR which brings the value into equivalence with the is more appropriate for the assessment of such organic matter is
FAMM equivalent YR. If the corrected VR and the FAMM given by the relationship
equivalent VR values are not the same within 0.1%, this usually HI. = 100 (S , - 0.26R) / TOe - R , where R = TOe 00 %11 100
indicates that the vitrinite population has not been accurately and %1 is the volume % inertinite in the organic matter' .
defined in one, or both, methods and the determinations should Recasting HI values may result in what appears to be a change
be repeated with this possibility in mind. This test is routinely in the identification of the kerogen type. Furthermore, organic
applied in TM studies involving FAMM and VR carried out in matter with a high inertinite content will have its T m x value
the CSIRO Division of Petroleum Resources and a near equal controlled by the liptinite and vitrinite components because of
ity of corrected measured VR and FAMM equivalent VR values their high volatile content relative to inertinite. T m" values will,
is regarded as strong evidence that both techniques are perform therefore, be lower than expected on the basis of the uncor
ing well and are not affected by oxidation, for example. For rected HI values of the organic matter. It is important that this
organic matter from the Jurassic and Cretaceous rocks of the possibility is kept in mind in North West Shelf TM studies,
North West Shelf, although the measured and FAMM equiva because inertinite-rich organic matter is very common, espe
lent VRs are commonly numerically different, they are consis cially in the important marine sequences.
tent with one another by reason of the suppression correction.
FAMM and VR results for Triassic samples are more com- 2 This equation is misprinted in Horstma n (1994).

EqVR (%) 5 EqVR (%)


5
A Coals B .................................... .OQM .. 0.40
0.40

'"
~

o ~
C
.........................................,. T ........;;\Q
c::i
~
............... 0 .50 r
~:)
.. ," '"'
o
j,\O ............... 0.50
:;:::;
, ,,.
~ :
I I ~.

c 0;0.~g18 , ~

o ,: ,/ .0. 33,
: / / c::J.:;.\O
>r
-
:;:::; ....................................... -: . .. -/ ....../ ........... ;I-~. 0.60
,,
en 0 .60
,, ,,
................................ r .,~:39 /

k
~
Q)

45.l3~:37,., 038;.<:>~\Q 0.70


I I 0\0

ro
r . .".. / .,//
0'35',0.36 ;~0 32,,'? ............................ :0
Q) 0 .70
'0:30 /
'044
: 05;1.:,0
.,.
u ~ /.0.39/ ,,' : PA7'
c ................................. ...... P6~'Q .5~ ..
I ,0.33 /

9 0.80
. ... I t- - - --. !/~

:.0.4 " 0.80 :0.49/. ',9 0


Q)
u Jurassic ' , , ,
1 ._ 0.43
en : 0.65.:. 0.39
'normal' ~ I I /

:;~56i 947
~ .... : ... ) ... / ;' Cf:rf
o v/trinHeJine 0.90
, : .,0.53
0.90

u::
:::l , I I ,
.0.37
: /
.... : t "-; .. / "),... "
I
"
/
I
1.00 ... , .. {- . f... .,'. ... ./. 1.00
,, , ! ' / / ,,'
,..
......................................... ~
1.10 ~ .0.58 1.10
: .1 .12 :
1.20 1.20
0.5
10 3 10' 10 2 10 3 10'
Fluorescence intensity at 400 s Fluorescence intensity at 400 s

Figure 2. A FAMM fluorescence alteration diagram on which the ratio of the final (700 s) to the initial fluorescence
intensity is plotted against the final fluorescence intensity derived from maceral fluorescence alteration curves. The
diagram shows the maturation pathway for orthohydrous telovitrinites (the 'normal' vitrinite line) and generalised
suppression iso-correction curves indicating the correction to be added to measured VR for the suppression effect. A and
B ar~ the average values of vitrinite populations with the same equivalent VR (0.7%), but with measured reflectances of
0.7% and 0.5% respectively.
70 R.W.T. WILKINS

460 Australian Permian Nonnally, maturation modelling of Rock-Eval data is car


ried out directly, using pyrolysis kinetics (Tissot & Espitalie
450 1975). The alternative of transforming T mox to equivalent VR
can only be done with much uncertainty and great caution
440
0- (Whelan & Thompson-Rizer 1993). Even among samples of a
~ defined type-for example, type III vitrinite-rich coals-changes
X 430
ro of maceral composition with age and geographical occurrence
E
I- have an important effect on equivalent VR (Powell et al. 1991).
420
This is illustrated in Figure 3, where a T max of 430C for a Ter
410
Netherlands Carboniferous tiary perhydrous Indonesian coal corresponds to an equivalent
'deviating' Netherlands Carboniferous
'normal' VR of little more than 0.5% and the same Tmax value for an
400 orthohydrous Netherlands Carboniferous coal gives an equiva
0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 lent VR of 0.75% .
VR(%) Regarding the question of the effect on T max of samples
Figure 3. Plot of T m" vs VR for Type III organic matter of showing VR suppression, Newman et al. (1994) found in a
different ages. (Data from Veld et aI. 1993, Powell et aI. study of serial plies of coal seams that there was a general
1991, and Teerman pers. comm. 1995). lowering of Tonax by up to 15C in samples with the lowest
reflectance (Fig. 4). Similarly, the Tmax values of Canadian Creta
ceous coals with high HI values have been noted to be anoma
470
lously low and the effect has been termed 'suppression of Tonax'
(Snowdon 1995). It should also be noted that contamination by
460 cavings and reworked organic matter cause anomalies which are
difficult to resolve without visual examination of samples.
450
u
o
-;:; 440
North West Shelf wells
co
E TM data from published and unpublished sources for three
f-
430 wells from the North West Shelf of Australia (Bowers-I,
Carnarvon Basin ; Flamingo-I, Bonaparte Basin; Kalyptea-l,
420 Browse Basin) are discussed in this section .

410
Bowers-l
0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 Figure 5A shows depth vs VR and FAMM equivalent VR pro
VR(%) files for Bowers-l based on CSIRO data. In Figure 5B these
Figure 4. Relationship between VR and T for serial plies data are compared with depth vs VR and SCI equivalent VR
from several New Zealand Tertiary coa"i"seams, showing profiles based on RRI results. The results from all three tech
the effect of perhydrous vitrinite compositions (with VR niques and both laboratories are broadly equivalent below the
suppression) on T m" of iso-rank samples. Samples from the major unconformity at 3700 m, which separates the marine
same coal seam are linked (after Newman et al. 1994). Middle Jurassic and fluviodeltaic Triassic sediments. Above the

A Bowers-1 B
FAMM equiv. VR % o VR % (CSIRO) SCI equiv. VR % 0 VR % (RRI)
2000 o 2000

2500 2500
0

0

--S
o
0


-
- 3000
0
3000

-
E 0

..r:::.
0..
..r:::. 0

Q)
Q.
Q)
0

o 3500 0

0 3500
0
0
0

0
De
4000 4000 0

0
0 0

4500 L-_~~~~~~-'---_ _~ 4500


0.2 2 0.2 2
VR% VR%
Figure 5. Profiles for the Carnarvon Basin well Bowers-l of (A) depth vs VR and FAMM equivalent VR, based on
CSIRO data from Ellacott et al. (1992b), and (B) depth vs VR and SCI equivalent VR, based on data from Robertson
Research Australia (1988).
THERMAL MATURITY INDICATORS 71

unconfonnity, FAMM and SCI equivalent VRs, and VRs from The discrepancy between the two sets ofVR results can be
RRI are in broad agreement, but they are in clear disagreement investigated using the systematic relationships known to exist
with the CSIRO VR results. It is also worth recording that the between VR and FAMM (Fig. 2). Perhydrous vitrinite
CSIRO VR and FAMM studies were carried out independently populations were revealed in the fluorescence alteration diagrams
by different workers using the same materials. of all samples above the major unconformity. Figure 6 shows

Bowers-1

o Inertinite & liptinite


- Vitrinite

10 10
2955-60 m 3060-65 m

EqVR% EqVR %
.
0.4
0.4
5 5 ~Q

~
c
o 0 0\0
+=' c v .......... " 0.5

~ ~ c
~
0\0
c
o
C
0
+=' .................... y..
c '"0.6
~ ~
2
co
2
co 0.7
Cll Cll
u ()
c
c
0 8
Cll
u
Cll
()
en

en
~ ~
o 0
::J ....... 0 ..... 1.0
::J
u:::: ......0 ... . ......... .0 ...... 0 .............................. 0 ... ~1.1 u::::
.....
1.1

1.2 1.2

10 4
Fluorescence intensity at 700 s Fluorescence intensity at 700 s
Figure 6. Fluorescence alteration diagrams for two samples from the Flacourt Formation of Bowers-l
(data from Ellacott et al. 1992b). Filled circles, vitrinite; unfilled circles, inertinite and Iiptinite.

Flamingo-1
1500
FAMM equiv. VR%
o Meas'd. VR%
+ SCI equiv. VR%

2000

1.60 Carnarvon Basin


-
2500
o Flamingo-1
--
-
S 1.40
0

---
~
0 ..
Q)
1.20
0
3000 ..--. 1.00
~
0

-- --
-i---~-
'-'

0::: 0.80
> El~'-
3500
0.60 Elo
DO_ -,_
:-

0.40
- -I- -~_
~~
0.20
4000
0.2 2 0.00
Vitrinite Reflectance (%) 0 2 4 6 8 10
Figure 7. Profiles of depth vs VR and equivalent VR from FAMM SCI
and SCI for Flamingo-I. All measurements based on nOM with
the exception ofthe the sample from 3631 m, which contains vitrite Figure 8. Plot ofVR vs SCI, showing results on the Bonaparte Basin
stringers (after Wilkins et al. 1992a, Robertson Research Australia well Flamingo-l compared with a selection of data from 24
1988; KK data in Botten & Wulff 1990). Carnarvon Basin wells (Robertson Research Australia,1988).
72 R.W.T. WILKINS

fluorescence alteration diagrams oftwo samples from the Early numerical agreement with the FAMM equivalent VR values.
Cretaceous Flacourt F ormation in Bowers-I . For both samples, This may be a result of the difficulty in identifying the indig
a VR suppression of approximately 0.4% absolute is predicted. enous vitrinite population in kerogen concentrates. It is pos
This is close to the observed differences between measured VR sible that a population of reflectance-suppressed inertinite
and FAMM equivalent VR for both samples. Thus, although (Wilkins et al. 1994a) was followed in successive samples down
the CSIRO FAMM and VR results are numerically different, the well. The SCI results on Bowers-I support the observation
the values are nevertheless broadly consistent. Although the of Schwab et al.(l994) that spore colouration is not influenced
same test cannot strictly be applied to the RRI VR results be by the factors causing suppression of reflectance in vitrinite.
cause samples from the same depths have not been studied by
the FAMM technique, it seems most unlikely that all the samples Flamingo-l
above the unconformity could have contained orthohydrous As for Bowers-I , there exist VR results from two laboratories
vitrinites, given that vitrinites with perhydrous compositions (RRI and KK), as well as equivalent VR data from SCI (RRI)
were found in all CSIRO samples from this section of the well. and FAMM (CSIRO). Both sets of measured VR results give
It is concluded that the RRI VR results on Bowers-I are almost what is effectively the same profile on the depth vs VR plot
certainly in error, despite the fact they are in approximate (Fig. 7) and in this diagram, the combined measured VR data

F/amingo-1

10 10
Core 5-1 3329.3 m Core 6-2 3631.08 m

EqVR % EqVR %

0.4 0.4
5 ::.~ 5
0 0
""~ ""~ 0.5
c c
0 0
""~ ""~ 0.6
2 2
ro ro
Q) Q) 0.7
u u
c c
Q) Q)
u u 0.8
(/) (/)

E! E! 0.9
0 0
:oJ :oJ
u::: u::: 1.0
1.1
1.2
0
0.5 0.5
10 2 10 3 10 4 10 5 10 2 10 3 10 4 10 5
Fluorescence intensity at 700 s Fluorescence intensity at 700 s

Figure 9. Fluorescence alteration diagrams for two samples from Flamingo-I.

Ka/yptea-1

0 0
o Meas'd VR%
o Cavings
FAMM eqiv. VR%

1000 1000

-
I..r::: 2000

-
I..r::: 2000

C
0..
<II
3000

C
0..
<II
3000

"
".
4000
" 4000

.. -
..
5000 5000
400 420 440 460 480 500 0.2 2
T max (OC) Vitrinite reflectance (%)

Figure 10. Plots of depth vs T ,and measured and FAMM equivalent VR for the Browse Basin well
Kalyptea-l (from data given i~n'Ellacott 1992a).
THERMAL MATURITY INDICATORS 73

have been fitted with one line. Transforming the SCI data to VR The high liptinite content of the Kalyptea-l samples and
equivalent, using the RRI North West Shelf calibration their high liptinite to vitrinite ratios suggests that these samples
(Robertson Research Australia 1988), it is seen that the SCI are likely to exhibit VR suppression, and this is confirmed by
equivalent VRs are systematically displaced to lower values the high fluorescence of the vitrinite throughout the depth of the
than measured VR on this diagram. On the other hand, the well. Although the suppression correction is variable, an aver
FAMM equivalent VRs have systematically higher values than age suppression of about 0.3% is suggested by the FAMM data
the measured VRs. and this is in fair agreement with the observed differences be
On a VR vs SCI cross plot (Fig. 8), the RRI Flamingo-l data tween the VR and FAMM results. In summary, the results from
are compared with a selection of data from 24 Carnarvon Basin all three techniques are reasonably consistent.
wells (Robertson Research Australia 1988). For this basin, VR
and SCI are highly correlated. Most of the Flamingo-I data for
VR < 0.7%, however, plot off the trend of this line towards Discussion and conclusions
lower values of SCI for a given YR. The apparent inconsistency Some inconsistencies between different sets of TM data are
between the VR and SCI results for Flamingo-I , suggests that caused by technical errors in one or more of the techniques,
different calibration curves relating these parameters may be such as the inaccurate identification of the indigenous vitrinite
required for the Carnarvon and Bonaparte Basins. population for VR, difficulties in recognising oxidised samples
It is next necessary to investigate whether the FAMM
equivalent VRs are consistent with the measured VR results. o
With the exception of the core sample at 3631.08 m, which
contains vitrinite bands, all samples are core or cuttings contain
1000 Lip tin ite
ing only dispersed organic matter (DOM) . Two lines of evi
dence show that VR and FAMM results are broadly consistent.
E
-
-
Firstly, where the fluorescence alteration diagrams contain tightly 2000


constrained populations of vitrinite as in the examples of
Figure 9, they correctly predict the magnitude ofVR suppres .s:: 3000 ...
sion. For core 5-1 (3329.3 m) the predicted difference (0.35%) c. ...
(1) ......
between measured VR (RRJ and KK data; 0.6%) and FAMM
c 4000
......... ~...
equivalent VR (0.95%) is the same as the actual value . Core 6-2
(3631.08 m) is of particular interest, because the coaly bands
allow unequivocal identification of vitrinite. The reflectance of 5000 0
the vitrinite in this sample has been verified in CS1RO and other 20 40 60 80
laboratories and the value is always close to 1.0% (Wilkins et al.
1992a). FAMM equivalent VR for this sample is 1.2% and the
Maceral vol. (%)
predicted suppression of VR is about 0.2%, which is in agree
ment with the measured YR. Both sets of measured VRs and 0

--
FAMM equivalent VRs are broadly consistent, though the VR
and FAMM results differ numerically, but all are inconsistent 1000
Vitrinite

-
with the SCI equivalent VRs, either by reason of errors in the
SCI measurements or in their transformation to equivalent YR. E
2000
.s::
Kalyptea-l
c(1). 3000
Three sets of maturity data for Kalyptea-l ; VR, FAMM and
C
T are shown in Figure 10. The FAMM and VR studies were

c;;;ied out on the same samples by CSIRO and KK respec
tively. A preliminary examination of the VR, Tmax and HI data
suggests that the organic matter has not reached peak maturity
4000
....
rJI.

for oil generation at maximum depth (VR < 0.7%; T max< 445C). 5000 0 20 40 60 80
In addition, HI values, which are all below 200 mg / g TOC are
disappointing in terms of petroleum prospectivity. Maceral vol. (%)
However, the maceral analyses (Fig. 11) show that the or
ganic matter is mainly composed of inertinite (40~80 % ) and 0
liptinite (10~65%) , whereas the vitrinite in most of the samples

--
is <20%. The result of adjusting the HI values for the high
1000
Inertinite
inertinite content by Horstman's (1994) method is shown in

-
Figure 12 . This 'upgrades ' the assessment of generative poten
tial of the organic matter so that much of it has the aspect of
E 2000
Type II~III. As the classification ofthe organic matter is modi .s::
c(1). 3000
.-
fied from Type III towards Type II, with strong marine affinity
to the source of the liptinite, the expected Tmaxvalues would be
c
lower (Tissot et al. 1987). The Tmax value of marine Type II
4000
organic matter is approximately 450C at peak generation
(Espitalie et al. 1985), which is close to the values recorded

from about 4000 m in the well; thus the apparent inconsistency
between the FAMM equivalent VR and T max is largely resolved. 5000 0 20 40 60 80
Further support for the FAMM equivalent VR value is pro
vided from chemical maturity parameters based on sterane iso
Maceral vol. (%)
mers (Peters & Moldowan 1993). These have reached Figure 11. Variation of liptinite, vitrinite and
equilibrium at 4250 m, suggesting equivalent VR values > 0.9% inertinite maceral proportions with depth for
(C. Boreham, AGSO, pers. comm. 1996). Kalyptea-I (based on KK data in Ellacott I992a).
74 R. W. T. WILKINS

800 800 Although the technical and methodological problems of each


A B TM indicator can give rise to errors, many of the more interest
700 700
ing and systematic inconsistencies, however, may be only ap
600 600 parent. The terms rank and thermal maturity are used ,
II

II
respectively, to refer to the degree of coalification of coals and
..
500 500
the level of organic metamorphism of source rocks, but insofar
J: 400 J: 400
as some coals may also be source rocks, the terms are often used
300 300 interchangeably. However, Suggate (1990) pointed out that the
concepts of rank and maturity are not synonymous for coals if
their initial chemistry is different. Suggate's (1959) parameter
.*
200 200

III III Rank(S) is an attempt to cope with problems produced by dif


100
100
ferences in the chemistry of coals due to type, which originate
0
0 100 200 300 400 500
0
0 100 200 300 400 500
in differences in original plant material and degree ofhumifica
01 01 tion and gelification from the peat stage. The problem has been
further discussed by Sykes et al. (1992). Newman et al. (1994)
Figure 12. HI-OI plots of Rock-Eval data for Kalyptea-l (Analabs
data in Ellacott 1992a). A, raw data. B, HI adjusted for high succinctly described the relationship by distinguishing coal 'rank'
inertinite content by the method of Horstman (1994). as relating specifically to thermal history, whereas 'maturity'
refers to the chemical state achieved in response to both depo
sitional and burial history.
in FAMM analysis, or drilling mud contamination of cuttings Rank(S) is derived from routine coal analyses and the method
and sidewall core samples in Rock-Eval pyrolysis. Such errors is not readily applied to source rocks containing DOM; never
are difficult to identify unless duplicate determinations by the theless, such materials present the same problems as Suggate
same technique from different laboratories or, at least, by dif (1959) addressed for coals. The differences between the con
ferent operators in the same laboratory are in disagreement. cepts of , rank ' and 'thermal maturity' are well illustrated by the
At the methodological level , each ofthe TM indicators dis relationship between FAMM and VR in Figure 2, which is based
cussed in the examples above has its problems. Curiale et al. largely on data obtained from North West Shelf samples. The
(1989) noted that all kerogen-based maturation parameters de figure shows the average values of two vitrinite populations
pend for their efficacy on the degree to which the type of or with the same equivalent VR values, but with different mea
ganic matter being studied can be calibrated and allowed for. sured VR values. For the population plotting on the 'normal'
This comment is especially apposite to Rock-Eval Tmox, which vitrinite line, the VR and equivalent VR values are the same,
has little value as a maturity indicator unless the maceral com whereas for the second population, the values differ. Both popu
position can be identified with one of the organic matter types lations have the same rank, expressed by the equivalent VR
(r, n or III) defined by Espitalie et al. ( 1977). Maceral analyses results, but they have a different degree of thermal maturity
provide some of the most significant information for the assess expressed by the different VR results. FAMM is primarily a
ment of source rocks. Yet, for North West Shelf wells, only the rank indicator because, within the general limits of Type II and
relatively few samples submitted for petrographic examination Type III organic matter, it is insensitive to the hydrogen content
will have had their maceral composition estimated. Ifno direct of vitrinite. The effectiveness ofFAMM in extracting rank in
information on the maceral composition is available, attempts formation from Indonesian coals with a wide range of
can be made to estimate it from other geochemical data. The perhydrous and subhydrous compositions (Teerman et al. 1995)
difficulty of successfully carrying out this type of indirect de gives confidence in this approach to rank determination. By
termination is well illustrated by the example of Kalyptea-l contrast, while VR always functions as a TM indicator, it is
described above. Some North West Shelf organic matter, nota only a rank indicator when vitrinite has an orthohydrous com
bly the common association of inertinite and liptinite in marine position.
sediments, does not compare well with any ofthe classic types, The modelling ofVR by the method of Burnham & Sweeney
and methods of treating Rock-Eval data on such materials for (1989) is best described as rank-based, because the Carbonifer
organic matter typing and thermal maturity estimation are still ous coals providing most of the chemical data linking the py
being explored (e.g. Horstman 1994). rolysis equations to VR are likely to be close to orthohydrous in
Both SCI and VR techniques attempt to overcome the prob composition. Thus, thermal maturation modelling by this method
lems of organic matter type by restricting measurements to spe is only strictly valid when the vitrinites fro m a modelled well
cific macerals. Some methodological problems of both approaches are also orthohydrous. For sequences containing perhydrous or
have already been mentioned. Spore colouration appears to have subhydrous vitrinite compositions, FAMM equivalent VRs
the important advantage over VR in that it is unaffected by approximate VR values for orthohydrous vitrinites of the same
factors giving rise to reflectance suppression of vitrinite, though rank. Until appropriate thermal maturation kinetics for
more evidence needs to be accumulated on this point. On the perhydrous vitrinite compositions are available, better agree
other hand, the low resolution of the method and the consider ment between measured and calculated VRs in maturation mod
able uncertainties in transforming SCI to equivalent VR lessen elling is likely to be obtained using the FAMM equivalent VR
its value for thermal maturity modelling. The FAMM technique values rather than the measured VRs (Wilkins et al. 1994b).
accomodates changes in maceral composition because the si Regarding discrepancies from simple constant heat flow matu
multaneous acquisition of both maturity and compositional in ration models, the possibility must also be kept in mind that
formation enables the effects to be separated. Nevertheless, it many North West Shelf wells are drilled close to faults which
too has methodological problems such as the restricted range have been significant conduits for fluids from depth over signifi
(equivalent VR = 0.4- 1.2%), and errors of 0.1% absolute can cant periods of time (O ' Brien 1995), and the effect of such fluid
easily occur near the limits of this range (Lo et al. 1996). Such flow on the thermal maturation of the organic matter is poorly
disadvantages are balanced by the highly systematic relation known at present.
ship between VR and FAMM equivalent VR, which accommo Two final comments may be made. Firstly, the often quoted
dates the VR suppression effect. As has been shown for the opinion- that the more independent TM parameters that agree,
three North West Shelf well examples, this can be used to check the greater the likelihood that the indication ofthermal maturity
that the indigenous vitrinite population in both the FAMM and is correct-could be misleading. If, as Price & Barker (1985)
VR methods has been accurately identified, and it increases the maintain, all maturation indices are retarded for hydrogen-rich
confidence in both results. kerogens, care must be taken not to confuse the strong indica-
THERMAL MATURITY INDICATORS 75

tion ofthennal maturity with rank. Secondly, it should be stressed Horstman, E.L., 1994. The effect of inertinite on kerogen ap
that TM indicators do not have to agree. This is an inevitable praisal by programmed pyrolysis, North West Shelf, Aus
consequence of the range of activation energies amongst the tralia. Australian Petroleum Exploration Association Journal,
wide variety of reactions on which the indicators are based and 34,297- 306.
the differing thennal histories of sedimentary basins. Thus, it Jones, R.W. & Edison, T.A., 1978. Microscopic observations
can not be expected that there will be any simple scheme of of kerogen related to geochemical parameters with emphasis
correlation between TM indicators (Heroux et al. 1979, Tissot on thermal maturation. In: Oltz, D.F. (editor), Low tem
et al. 1987). perature metamorphism of kerogen and clay minerals. Los
Angeles: Pacific Section , Society of Economic
Acknowledgements Paleontologists and Mineralogists, 1- 12.
Lo, H.B ., Wilkins, R.W.T., Ellacott, M.Y. & Buckingham, C.P. ,
I am greatly indebted to Mike Ellacott who carried out many of 1996. Assessing the maturity of coals and other rocks from
the FAMM studies referred to in the paper, and Carol North America using the fluorescence alteration of multiple
Buckingham who reviewed all the data and constructed the dia macerals (FAMM) technique. International Journal of Coal
grams. I wish also to thank Nigel Russell, Neil Sherwood, Geology, in press.
Mohinudeen Faiz, Chris Boreham and Bob Alexander for their McCulloh, T.H. & Naeser, N.D., 1989. Thennal history of sed i
constructive criticisms of the paper. mentary basins: Introduction and overview. In: Naeser, N.D.
and McCulloh, T.H. (editors), Thermal history of sedimen
References tary basins. Methods and case histories. Springer Verlag,
Botten, P. & Wulff, K , 1990. Exploration potential of the Timor New York, 1- 12.
Gap Zone of Co-operation. Australian Petroleum Explor Newman, J., Price, L.C. & 10hnston, 1.H., 1994. Variations in
ation Association Journal, 30, 68- 90. source potential and maturation of New Zealand coals, based
Burnham, A.K. & Sweeney, J.J. , 1989. A chemical kinetic model on relationships between conventional coal chemistry, Rock
of vitrinite maturation and reflectance . Geochimica et Eval pyrolysis, and GCMS biomarkers. 1994 New Zealand
Cosmochimica Acta 53 , 2649- 57. Petroleum Conference Proceedings, 370- 384.
Cooper, B.S., 1977. Estimation of maximum temperatures at O'Brien, G.w., 1995. AGSO's 'fill-spill ' project: reducing ex
tained in sedimentary rocks. Tn: Hobson, G.D. (editors), ploration risk on the North West Shelf. Australian Geolog
Developments in petroleum geology- I. Applied Science ical Survey Organisation, Research Newsletter, 23, 10- 11.
Publishers, London, 127- 146. Peters, KE., Ishiwatari, R. & Kaplan, I.R., 1977. Color of kerogen
Curiale, J.A., Larter, S.R., Sweeney, R.E. & Bromley, B.W., as index of organic maturity. American Association of Petro
1989. Molecular thennal maturity indicators in oil and gas leum Geologists Bulletin, 61 , 504- 510.
source rocks. In: Naeser, N.D. & McCulloh, T.H. (editors), Petersen, N.F. & Hickey, P.J., 1985. Visual kerogen assessment
Thennal history of sedimentary basins. Methods and case ofthennal history. American Association of Petroleum Geol
histories. Springer Verlag, New York, 53- 72. ogists Bulletin, 69, 296.
Diessel , C.F.K., 1992 . Coal bearing depositional systems . Powell, T.G. , Boreham, C.J., Smyth, M., Russell, N. & Cook,
Springer Verlag, Berlin. A.C. , 1991. Petroleum source rock assessment in non-marine
Ellacott, M.Y. , Russell, N .J. , Buckingham, C.P., Wilmshurst, sequences : pyrolysis and petrographic analysis of Austral
J.R. & Wilkins, R.W.T. , 1992a. Fluorescence alteration and ian coals and carbonaceous shales. Organic Geochemistry,
suppression of vitrinite reflectance. Report 4: Kalyptea-1. 17,375- 394.
CSIRO Division of Exploration Geoscience, Restricted In Price, L.C. & Barker, C.E., 1985. Suppression of vitrinite re
vestigation Report 295R, 1- 78. flectance in amorphous rich kerogen- a major unrecognised
Ellacott, M .Y. , Russell , N.J., Buckingham, C.P., Wilmshurst, problem. Journal of Petroleum Geology, 8, 59- 84.
J.R. & Wilkins, R.W.T., 1992b. Fluorescence alteration and Robertson Research Australia, 1988. Oil and gas fields of
suppression of vitrinite reflectance. Report 7: Bowers-I. Australasia. Robertson Research Australia Pty. Ltd., Sydney.
CSIRO Division of Exploration Geoscience, Restricted Quick, J .c., 1994. Iso-rank variation of vitrinite reflectance and
Investigation Report 313R, 1- 56. fluorescence intensity. Tn: Mukhopadhyay, P.K. & Dow,
Ellacott, M.Y., Russell, N.J. & Wilkins, R.W.T. , 1994. Trouble w.G. (editors), Vitrinite reflectance as a maturity parameter.
shooting vitrinite reflectance problems using FAMM. A American Chemical Society, Symposium Series, 570, 64-
Gippsland and Otway Basin case study. Australian Petro 75 .
leum Exploration Association Journal, 34, 216-230. Schwab, KW., Smith, M.A . & van Gijzel , P., 1994. Electronic
Espitalie, J. , Deroo, G. & Marquis, F., 1985. La pyrolyse Rock color measurements of palynomorphs: a better method for
Eval et ses applications. Revue de I'lnstitute Franyais du defining thermal maturity in organic-rich sediments without
petrole, 40,.755- 784 . the problems of suppression. Society of Organic Petrology,
Espitalie, J. , Laporte. J.L. , Madec, M. , Marquis, F. , Leplat, P. , 12th Annual Meeting, Abstracts and Program, 3- 5.
Paulet, J. & Boutefeu , A. , 1977 . Methode rapide de Senftle, J.T., Landis, C.R. & McLaughlin, R., 1993 . Organic
caracterisation des roches meres, de leur potentiel petrolier petrographic approach to kerogen characterization. In: Engel,
et de leur degree d' evolution. Revue de l' lnstitute Franyais M .H. & Macko , S.A . (editors), Organic geochemistry.
du Petro Ie. 32, 23-43 Plenum Press, New York, 355- 374.
Hao Fang & Chen Jianyu, 1992. The cause and mechanism of Peters, KE. & Mo1dowan, 1.M., 1993 The biomarker guide:
vitrinite reflectance anomalies. Journal of Petroleum Geol interpreting molecular fossils in petroleum and ancient
ogy 15, 419-434. sediments. Prentice Hall, New Jersey.
Haseldonckx, P., 1977. Relation of palynomorph colour and Stach, E., Mackovsky, M.Th. , Teichmliller, M ., Taylor, G.H. ,
sedimentary organic matter to thennal maturation and hydro Chandra, D. & Teichmliller, R., 1975. Coal petrology. 2nd
carbon generating potential. In : Generation and maturation edition. Borntraeger, Berlin.
of hydrocarbons in sedimentary basins. CCOP Seminar, Sept. Staplin, F.L., 1977. Interpretation ofthermal history from colour
1977,41 - 53. of particulate organic matter. Palynology, 1, 9- 18.
Heroux, Y. , Chagnon, A. & Bertrand, R., 1979. Compilation and Staplin, F.L., 1982. Determination of Thermal Alteration Index
correlation of major thermal maturation indicators. Ameri from the from the colour of exinite (pollen, spores). In:
can Association of Petroleum Geologists Bulletin, 63, 2128- Staplin et al. (editors), How to assess maturation
2144. paleotemperatures. SEPM Short Course Number 7. Society
76 RW.T. WILKINS

of Economic Paleontologists and Mineralogists, Tulsa, 7-9. Tissot, B.P., Pelet, R. & Ungerer, P. , 1987. Thermal history of
Suggate, R.P., 1959. New Zealand coals: their geological setting sedimentary basins, maturation indices, and kinetics of oil
and its influence on its properties. New Zealand Depart and gas generation. American Association of Petroleum Geol
ment of Scientific and Industrial Research, Bulletin 134, 1- ogists Bulletin, 71 , 1445- 1466.
113. Van Krevelen, D. w., 1993. Coal: typology- chemistry- physics
Suggate, R.P., 1990. Variability in Type III organic matter at the constitution. Elsevier, Amsterdam .
initiation of diagenesis. In: Nuccio, Y.F. & Barker, C.E. Veld, H., Fermont, W.J.J. & Jegers, L.F., 1993 . Organic petro
(editors), Applications of thermal maturity studies to en logical characterization of Westphalian coals from The Neth
ergy exploration. Rocky Mountains Section, SEPM. , 45- erlands: correlation between Tmax, vitrinite reflectance and
52. hydrogen index . Organic Geochemistry, 20, 659- 675.
Suzuki , N., Matsubayashi, H. & Waples, D.W., 1993. A simpler Whelan, lK. & Thompson-Rizer, C.L., 1993. Chemical methods
kinetic model of vitrinite reflectance. American Association for assessing kerogen and protokerogen types and maturity.
of Petroleum Geologists Bulletin, 77 , 1502- 1508. In: Engel , M.H. & Macko , S.A. (editors) , Organic
Sweeney, 1.1. & Burnham, A.K., 1990. Evaluation of a simple Geochemistry. Plenum Press, New York, 289- 353 .
model of vitrinite reflectance based on chemical kinetics. Wilkins, R.W.T. , Ellacott, M.Y. , Wilmshurst, J.R. & Bucking
American Association of Petroleum Geologists Bulletin, 74, ham, C.P., I 994a. The suppression of inertinite reflectance .
1559- 1570. Organic Geochemistry, 21 , 871 - 875.
Sykes, R., Suggate, R.P. & King, P.R ., 1992. Timing and depth Wilkins, R.W.T. , Russell, N.J . & Ellacott, M.Y. , 1994b. Fluo
of maturation in southern Taranaki Basin from reflectance rescence alteration and thermal maturation modelling of
and Rank(S). Proceedings, 1991 New Zealand Oil Explor Carnarvon Basin wells. In: Purcell, P.G . & R.R. (editors),
ation Conference, Christchurch, New Zealand, Ministry of The sedimentary basins of Western Australia. Proceedings
Commerce, 373- 389. of Petroleum Exploration Society of Australia Symposium,
Teerman S., Ellacott M.Y. , Wilmshurst J.R. & Wilkins R.W.T., Perth, 415-434.
1995. Interrelationships between FAMM, vitrinite reflect Wilkins, R.W.T., Wilmshurst, J. , Hladky, G., Ellacott, M.Y. and
ance and Rock-Eval results on Indonesian Tertiary coals. In: Buckingham , C. P. , 1992a. The suppression of vitrinite
Grimalt , 1.0 . & Dorronoso , C . (editors) , Organic reflectance in some North West Shelf wells: Barrow-I ,
geochemistry: developments and application to energy, Jupiter-l and Flamingo-I . Australian Petroleum Exploration
climate environment and human hi story. A.l.G .O .A., Association Journal, 32( I), 300- 312.
Donostia-San Sebastian, 466-468 . Wilkins, R.W.T., Wilmshurst, J.R., Russell , N.J. , Hladky, G.,
Ting, T.C. & Sitler, J.A., 1989. Comparative studies of th e Ellacott, M.Y. & Buckingham , C., 1992b. Fluorescence
reflectance of vitrinite and sporinite. Organic Geochemistry, alteration and the suppression of vitrinite reflectance. Organic
14, 247- 252 . Geochemistry, 18, 629- 640.
Tissot, B.P. & EspitaJie, J., 1975. L'evolution thermique de la Wilkins, R.W.T. , Wilmshurst J.R., Hladky, G., Ellacott, M.Y. &
matiere orga nique des sediments: applicati ons d' une simu Buckingham, C.P., 1995. Should flu orescence alteration re
lation mathematique. Revue de I' lnstitute Franyais du PetroIe place vitrinite reflectance as a maj or tool fo r thermal mat
30,743- 777. urity determin ati on in petroleum ex pl oration? Organic
Geochemistry, 22, 19 1- 209.

You might also like