Karena aku ngga bisa dateng aku sharing papernya Rob Simm yg berhub dengan
discovery of Buzzard ini ...
kalo ada yg dateng sharing donk ..
aku cuplik dikit dari kalimat dibawah :
"So, what of Buzzard Field?. It cannot be denied that the decision to
drill was seemingly based purely on a geological model, mass flow sands
dumped in a graben and pinching out up dip to the west. A risky model in
anybody's book, and many companies (most of which no longer exist) decided
to give the area a wide berth."
rdp
======
Amplitudes, Risk and all that Geophysical Malarkey
Rob Simm1, Andrew Armour2 and Martyn Millwood Hargrave2, 1Rock Physics
Associates Ltd, 2IKON Science Ltd
11th February 2003, Burlington House, Geological Society, London
PESGB London Evening Meeting
'There was no geophysical malarkey involved in finding the Buzzard field'
said Graham Dore in his PETEX presentation last December. He was right of
course as we shall discuss later, but 'malarkey', meaning humbug,
foolishness or nonsense, is an interesting, if not telling, choice of
word. We take it to mean that in this instance there was some satisfaction
in being spared the frustration and agony of incorporating geophysical
information into the decision to drill that on previous occasions has
proved confusing and even misleading. The loud cheer from the audience
showed that he had made an almost visceral connection with the gathered
brethren.
It points to a significant problem in the industry. The problem is that
although it is recognised that seismic may in certain situations show
effects that are related to the presence of hydrocarbon (and if they are
recognised as such they may be used to lower the perceived risk), the
practice of using seismic information as an input to drilling decisions is
not uniformly good or pleasant for all. In our opinion there is a good
deal of 'malarkey' going on, particularly when unrealistic claims are made
for the significance of amplitude information. Guarding against such
'malarkey' is the subject of this talk.
Amplitude (DHI) interpretation
The term 'amplitudes' is used here in a general sense to cover all
interpretation based on reflection seismic and its derivatives (so it
includes full and partial stack interpretation, AVO analysis and
(elastic) inversion) and attribute derivatives. In reflection seismic the
acronym DHI (Direct Hydrocarbon Indicator) is generally used within the
industry to denote an effect on reflection seismic data that can be
attributed to the presence of hydrocarbon. DHI's commonly comprise (but
are not limited to)
Single or associations of reflection signatures (pre and post stack,
bright spots dim spots, phase reversals) or impedance characteristics
linked to the effects of hydrocarbon via rock physics models
down dip limit/termination or structural conformance of amplitude
flat spots
The criteria for the use of the term DHI for a recognised seismic effect
are in fact quite stringent but are often applied recklessly or foolishly.
A clear consistency has to be shown between the observed attributes and a
rock physics model that illustrates a likely hydrocarbon interpretation.
In addition consistency between the signature and other expected effects
predicted by the model needs to be evaluated but very often are not. It is
too easy to neglect other plausible causes for the effects as they may not
lead to a viable prospect ? the human mind appears to be constantly
engaged in a creative search of corroborative evidence for its favoured
hypothesis.
There are a whole host of reasons why we can get the interpretation wrong.
These include:
non-uniqueness of the effects ? (ie other geological scenarios are
responsible, eg high porosity or low gas saturation)
seismic polarity is misinterpreted there are interpretive problems with
seismic data acquisition and/or processing
the model fails (ie the assumption that seismic can be approximated as the
convolution of a wavelet with a reflection series determined from
elastic/isotropic rock properties is wrong). Anisotropy plays a role in
giving a 'false positive' indicator. Given the practical problems of
parameterising anisotropic models we currently don't know enough about how
often this actually happens.
Amplitudes in the Risking Context
Clearly if a verifiable DHI is present on a prospect then it is possible
that the risk on the prospect may be considerably reduced relative to a
standard geological risk either by utilising the DHI evidence within a
probabilistic risking scheme or, possibly more dangerously, by using it to
override the risking scheme. There are pitfalls, however, at almost every
level of the process. One pitfall is that the DHI interpretation tends to
be invoked too readily, for example when there has been no play specific
corroboration through modelling or direct analogy of the effect(s) under
question. What is more, in these situations the term DHI implies more
certainty of hydrocarbon presence and a much narrower range of outcomes
than is warranted (Citron and Rose 2001). Loose thinking combined with big
promises can be a fatal combination!. Equally fatal is the confidence
trick where we are blinded by the elegance of the positive model, how can
it possibly fail?
There is always a risk in DHI interpretation in exploration (as some of us
know from bitter experience) and we must get away from the idea that if
DHI's really worked there would be no need for risking. Many companies and
individuals follow this 'silver bullet' idea and it is no surprise that
the DHI approach moves rapidly in and out of fashion. It is better to
think of the Casino analogy (Rose 1999), in which we hope to stack the
odds in our favour over a certain period of time with a portfolio that is
risked appropriately. We wont know necessarily which particular wells will
come in but over the life of the portfolio we believe that an appropriate
use of use of amplitude information will put us ahead. Companies that
cannot afford another dry hole should hope for lady luck.
The relationship between DHI's and exploration risk is neatly illustrated
by the classic example of the Yegua trend in the Gulf of Mexico. This is a
mature gas play comprising shallow high porosity sands that give bright
spots on stacked sections and increasing amplitude with offset on
pre-stack gathers (shale/brine sand reflections generally show opposite
polarity and decreasing amplitude with offset). 84 wells that were
drilled on 'AVO anomalies' were documented in a study by Allen et al
(1993). The commercial success rates improved dramatically, from around
5-10% to 50%, when AVO techniques were employed (using 2D seismic data).
Clearly the DHI's and associated risks described above are specific to the
Yegua play and it would be foolhardy to take these particular DHI's and
expect them to work in a similar way in different types of plays for
example West of Shetlands. However you could argue that on plays less
mature than the Yegua the likely chance of success associated with DHI's
must be considerab
ly less than 50%.
In the North Sea, most of the oil fields have been found without the
explorations being driven by amplitude information. There are numerous
examples of interpreters spotting the critical seismic DHI but
misinterpreting it. Later field studies bring to light the real
interpretation. What is certain is that amplitude technologies are now
adding considerable value in field development, owing to the high degree
of calibration available. On the basis of bottom line benefit Time-Lapse
Seismic techniques, underpinned by thorough rock physics, have become
established practice in many of the larger oil and gas companies within
the last 5-6 years.
Whilst the challenge is always there to use the lessons from the fields to
drive the exploration models in partially explored basins, in these
situations it is easy to convince ourselves that we know more than we
actually do. Very often if a well is available there is a tendency to
believe that it contains all that is necessary for calibration (including
all likely variability). Our problem then is to ask ourselves 'what is the
likelihood that the model will hold over the prospect area?'. In some
cases significant changes can occur to invalidate close well control
whilst in other situations a well 50km from the prospect may be entirely
relevant.
Calibration is clearly the key to the application of DHI's in risk. Our
models are only as good as the data on which they are based. In virgin
basins where there is stratigraphic uncertainty and no calibration it may
be completely unrealistic to try and apply DHI's in the risking process.
In such areas conventional AVO analysis can give us an idea of the degree
to which different responses are anomalous but it cant tell us how likely
a particular seismic effect will be related to hydrocarbon rather than for
example a high porosity brine sand (or any other as yet undefined
lithology combination for that matter). At best (and if you are lucky
enough to have a number of potential prospects) it can be used as a
ranking tool.
Assigning risk to DHI's
Very often we don't have the statistics to support the risks applied to
DHI's, usually because we are trying to apply the techniques in non-mature
areas. And when it comes down to it, whichever way you assign a risk
significance to DHI's it is highly subjective. Fundamentally it depends on
the level of knowledge of the play and an understanding of the particular
DHI characteristics of the target. With greater knowledge of the play the
vagaries of the DHI signature (and its relationship to factors such as
data quality) are likely to be more completely understood. Two companies
may recognise the same seismic effects but place different significance on
them simply because of differences in the understanding of the play. In
the Yegua example noted above 10 out of 84 prospects had a questionable
DHI (Allen et al 1993).
Many companies have attempted to formulate a way of incorporating
seemingly meaningful geophysical observations into risking systems to aid
the decision process. One approach is to use the seismic to give a chance
of success factor which can then be compared to a traditional risking
approach. A generic (and non-specialist) approach might be to use the risk
matrix idea of Citron and Rose (2001) but in which the axes are related to
the knowledge of the play (in terms of the seismic responses of lithology
and fluid, calibration if you like) and the confidence in the recognition
of the DHI(s). A questionnaire for each component would help arrive at the
appropriate position on the matrix. Chance of success values will be
specific to a particular play for a given range of analytical techniques.
Figure 1. An example of a chance of success matrix based on DHI's. Note
that each matrix is specific to a particular target in a specific play.
As well as evaluating the value of a DHI in the possible lowering of risk
we also need to address the significance of a lack of a DHI in a situation
where one would be expected (ie where we might use the lack of a DHI to
increase the risk). Interestingly, examples of 'false negatives' (ie
discoveries where there is no DHI but with the given data quality one
would be expected) are thin on the ground.
Buzzard and the idea of play development
So, what of Buzzard Field?. It cannot be denied that the decision to drill
was seemingly based purely on a geological model, mass flow sands dumped
in a graben and pinching out up dip to the west. A risky model in
anybody's book, and many companies (most of which no longer exist) decided
to give the area a wide berth. The seismic gives only general information
on stratigraphy as it is contaminated with multiples of different
varieties. Subsequent analysis of the pre-stack seismic, coupled with a
rock physics model based on the dry well down dip, perhaps surprisingly
does show attributes that appear to be related to the presence of
hydrocarbon, possibly even indicating a down dip limit for 2 prominent
sand prone intervals.
Whether or not the attributes described from Buzzard are convincing enough
to have been used reliably in a predictive manner prior to the Buzzard
discovery well or whether they provide a way of predicting further oil
pools in the same play is a question that many are now trying to answer.
Certainly given these results and the experiences from other North Sea
fields, there is a huge incentive to unravel the signature of the oil
filled sands and there is likely to be a role for calibrated amplitudes in
extracting even greater value from the Buzzard field development.
Ways forward with amplitude interpretation
Probably the most important element in guarding against geophysical
malarkey is a cross-discipline understanding of what amplitudes can and
can't do for us. Non-specialists (i.e. most geophysicists, geologists and
increasingly managers and engineers) need to be able to ask the right
questions to put the seismic information into context. This would help
communication and ensure that the geophysical interpretation is done in
the most rigorous way possible. This technology transfer can only happen
if:
There is a greater general understanding and access to seismic analysis
techniques linked to calibration methods, including rock physics, and
their limitations
There is access to a knowledge database of previous examples, the good,
the bad and the downright ugly.
Established training courses can go some way to achieving this, but
fundamentally there has to be an active mind-set within each company to
synthesise past experiences into 'learnings' that are made available to
those who need them. The value of the knowledge data base should not be
underestimated and it is never too late to begin the process of developing
it. What is not enough is for the geophysical priesthood to simply write
guidelines and prescribe expert -authorised workflows. They do not ensure
communication and in the worst cases can actually stop people thinking.
References
Allen, J.L, Peddy, C.P. and Fasnacht, T., 1993. Some AVO failures and what
(we think) we have learned. The Leading Edge, March.
Citron, G.P., and Rose, P.R., 2001. Challenges with amplitude-bearing,
multiple zone prospects. The Leading Edge, August.
Rose., P.R., 1999. Taking the risk out of petroleum exploration; the
adoption of systematic risk analysis by international corporations during
the 1990's. The Leading Edge, February.
"Titi Tabusalla" <[EMAIL PROTECTED]>
18/02/2003 08:43 AM
Please respond to fogri
To: <[EMAIL PROTECTED]>
cc:
Subject: [fogri] TODAY IPA TALK!!!
IPA PROFESSIONAL DIVISION
LUNCHTIME TALK
DATE : Tuesday, February 18, 2003
TIME : 11:45 Hrs.
VENUE: Golden Ballroom, Jakarta Hilton International
Subject: "Amplitudes, Risk and all that Geophysical Malarkey"
Speaker: Martyn Millwood Hargrave1, Dr. Rob Simm2 and Dr. Andrew Armour1
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
Visit FOGRI Website: http://fogri.or.id
FOGRI Archive: http://www.mail-archive.com/fogri%40iagi.or.id/
---------------------------------------------------------------------
