Forecasting future production from past discovery

Forecasting future production from past discovery
OPEC seminar
Jean Laherrere
Sept.28, 2001
OPEC and the global energy balance: towards a sustainable energy future
Vienna
Sept. 28-29, 2001
“Forecasting future production from past discovery”
Jean Laherrere
jean.laherrere@wanadoo.fr
-Abstract
There is a huge discrepancy between the “political” values of the reserves by country as
reported by Oil & Gas Journal, World Oil, BP Statistical Review, OPEC… and the
“technical” values which are confidential for most countries. Yet, most production forecasts
by official agencies are based on the political data. Some countries report minimum values
(e.g. the US with Proved values), others report maximum values (e.g. the FSU), and most
countries report likely or median (called Proven + Probable) values which are generally close
to, yet lower than, “mean”, e.g. “expected”, values. When technical data are used to calculate
the “mean” values of field reserves, a good fit is found between annual (and cumulative)
discoveries and annual (and cumulative) production, the former being close to the latter with a
time translation of a certain number of years. This procedure makes it possible to forecast
future production from the corresponding past discovery trend. Examples shown for
conventional oil are the US Lower 48, FSU, France, UK, Middle-East, deepwater and the
world outside “Middle-East & deepwater”, and for conventional gas, the North America.
A long-term forecast for World production of all hydrocarbons, based on these methods, is far
below all the scenarios developed for the 2000 Third Assessment report of the IPCC.
-Reliability of the data
Publishing reserves (even production) data is a political act and depends upon the desired
message: either optimistic for the financial community when trying to borrow money, or
cautious for the shareholders in order to leave room for later grow. This explains why political
(or financial) data are very different from the technical data, which are usually kept
confidential.
Usually reserves (which should represent the future production) are classified on the degree of
certainty. Proved (P or the minimum) is associated to a high probability level (90% or P90 in
the probabilistic approach of the SPE/WPC/AAPG rules) meaning that the reserves will be
larger than P in more than 90% of the possible outcomes. Proven + Probable (2P) could be
either the most likely value (mode), or the median value (50% probability or P50 or as likely
as unlikely) or the statistical “mean” (also called weighted average, or expected value) that
summarizes all the possible outcomes. Proven + Probable + Possible (3P or the maximum) is
associated to a low probability level (P10 that means that the probability that reserves might
be larger than 3P is less than 10%).
In the US, the practice of only reporting Proved reserves, omitting the Probable and Possible
reserves to comply with SEC (Securities & Exchange Commission) rules means that the
declared reserves are systematically under-reported, leaving room for future growth. During
the past 20 years, 88% of the annual additions of oil reserves come from the re-evaluation of
past discoveries because the previous estimates were systematically too conservative. An
1
OPEC seminar
Jean Laherrere
Sept.28, 2001
obvious advantage is that it permits the company to demonstrate a significant growth of past
reserves, even when no discovery was made, a possibility that oil and gas companies find to
their advantage. Conversely, the obvious drawback is that country data based on such underestimated individual reserves is so conservative that it has no meaning (called an “illegal
addition of proved reserves” by Capen 1996). Yet this is done every year by the oil journals
and many “experts”. Only the addition of “mean” field reserves corresponds with the “mean”
reserves of a country. A Monte Carlo simulation is necessary to obtain the P90 value for a
country from the P90 values of individual field reserves (and the rest of the probability
distribution).
Regarding OPEC, the production quota allocated to their member countries partly depends
upon their declared reserves. This led these countries to make large upward revisions of their
official reserves in the 80s despite of the absence of any exploratory work. There was
obviously no incentive to explore when it was not possible to produce at the full capacity of
the existing fields. OPEC countries (except the Neutral Zone) increased their previous reserve
figures by 300 Gb even though no significant discovery was made. It is difficult to know what
is the right “mean” value in these circumstances. The reserves of these countries are certainly
large but are unlikely to be as large as the new official figures, which are hardly unchanged
for the past 10 years despite the drawdown of what was produced.
Oil & Gas Journal (OGJ) reports before the end of the year what they call “Estimated proved
reserves” as of end of the year. The reported values are compiled from inquiries with the
government agencies, but they are either of purely “political”, or based on the sum of
individual field figures that systematically under-estimate expected values because of
methodological flaws. When no answer to the enquiry was received, as most studies are not
completed by year-end, the previous figures are often repeated without change, implausibly
implying that the amount of hydrocarbons produced was exactly equal to the net addition of
new reserves during the year. This was the case for 80 countries in the latest OGJ report.
OGJ Dec18, 2000
80 countries
25 countries
total
end of 2000
change from 1999
oil Gb gas Tcf oil Gb gas Tcf
586
4 025
0
0
443
1 253
12.4
132
1 028 5 278
12.4
132
change %
oil
gas
0
0
2.9 11.8
1.2 2.6
These values are repeated by other sources, such as the BP Statistical Review and the USDOE reports. World Oil delays publishing until the middle of the year in order to obtain the
estimates of the previous year, and correct the early release. They also moved in 1991 to
report Proven + Probable but reverted to Proven in 1996. Analysts either mistakenly accept
these published data, which are reproduced by BP and USDOE, as valid, or, if they know
them to be false, use them to mislead the general public and the policy makers. The IEA did
just that in its latest World Energy Outlook even though it had made a courageous attempt to
reveal the true position in 1998. The oil companies furthermore use these flawed public data
to improve their image with the investing public.
The “expected” values of the reserves can only be estimated from technical data that are
generally held confidential and only available through “scout” companies. The present
technical estimate is backdated to the year of discovery. But some corrections are necessary
for some countries.
2
OPEC seminar
Jean Laherrere
Sept.28, 2001
A good example is that of the reserves of the Former Soviet Union (FSU) which are taken for
granted by all the official western bodies. They ignore that Khalimov, who presented the
Russian classification in the 1979 WPC, made a statement in 1993, revealing that the FSU
estimates were “strongly exaggerated due to the inclusion of reserves and resources that are
neither reliable nor technologically nor economically viable”, because they were based on the
maximum theoretical recovery factor. In fact, the reported reserves were close to their 3P
values.
Figure 1 shows that the world’s remaining “ political” oil reserves, as reported, have been
rising steadily from 100 Gb in 1950 to 1000 Gb in 2000, which has misled some analysts into
proposing that they can continue to do so. By contrast, the “technical” data show a decline
since 1980, with a peak around 1200 Gb.
Figure 1: World’s oil remaining reserves from political and technical sources
World's oil remaining reserves from "political"
(API, BP, World Oil, Oil & Gas Journal, OPEC) &
technical sources
1300
1200
1100
1000
900
800
700
600
500
400
300
200
100
0
1950
?
API oil P current
BP Review oil P current
WO oil P current
OGJ oil P current
OPEC oil P current
technical data O+C 2P "backdated"
1960
1970
1980
1990
2000
2010
The world’s remaining “political” gas reserves (figure 2) shows also a steady rise from 1000
Tcf in 1965 to 5000 Tcf in 2000, whereas the “technical” data indicates a leveling at 6000 Tcf
since 1980
Figure 2: World’s gas remaining reserves from political and technical sources
3
OPEC seminar
Jean Laherrere
Sept.28, 2001
World remaining gas reserves from political and
technical sources
7000
6000
technical 2P backdated
OPEC P current
WO P current
OGJ P current
5000
4000
3000
2000
1000
0
1950
1960
1970
1980
1990
2000
-Correlation between production and shifted discovery
In 1956, King Hubbert made a forecast that US oil production would peak around 1970. He
was right. His famous bell-shaped curve is justified as a statistical aggregation of many
independent producing fields, but it does not work when exploration is discontinuous or for a
domain covering only a limited number of fields. The main mistake in Hubbert modeling is to
assume only one peak, when most of the countries display several peaks. The important
message from Hubbert’s work, which is often forgotten by economists, is that oil has to be
found before it can be produced. It means that, if no constraints, the production curve has to
mirror the discovery curve, so that if there are several discovery cycles, production will also
have several corresponding cycles.
The correlation between production and the shifted discovery curve, when based on technical
data giving the “mean” values, is usually quite good for countries producing at full capacity
and not subject to drastic economic changes. But the correlation is not good for the swing
producers, being the 5 countries of the Middle East: Saudi Arabia, Kuwait, Iran, Iraq and the
UAE, as their level of production is set on the basis of political decisions.
-US Lower 48 States
As US reserves are reported only as Proved values, it is necessary to adjust them upward to
equate with a “mean” value. To do that, we rely on the published data comprising: the 1990
USDOE/EIA-534 report “US oil and gas reserves by year of field discovery”; the new fields
discovery reported in the annual reports from 1989 to 1999; and the growth function
4
OPEC seminar
Jean Laherrere
Sept.28, 2001
calculated by the MMS (USDOI/Minerals Management Service) on the basis of the average
annual revisions of reserves of the about 1000 GOM fields over time. According to this
function, the initial Proved estimate of the discovery year needs, on the average, to be
multiplied by 4.5 to approach the “mean” reserve 50 years later. This model needs to be
improved.
With the backdating of the “mean” reserves at the time of the discovery, there is a good fit
between the discoveries shifted by 30 years and the productions. Based on the shifted
discoveries, one may infer what the future production will be for the next 30 years. However,
as mentioned earlier, there is evidence that a new oil cycle has started at the end of the 80s
with the coming on production of the deepwater discoveries.
The following figure 3 displays US Lower 48 annual oil production and annual discoveries
shifted by 30 years. Note the impacts of the depression of the 30s, the proration at the end of
the 50s and the high price in 1979. They explain the two “shoulders” of the production curve,
which for unknown reasons occurred at about the same level at 2.6 Gb/a.
Figure 3: US Lower 48 annual oil production and discovery shifted by 30 years
5
US (Lower 48): annual production and annual discovery
(Attanasi & Root AAPG 1994 & DOE annual reports
1995-99) grown with MMS model (*4.5) shifted by 30
years
mean discovery shifted 30 yr
oil production
production crude oil only
crude oil less deepwater
4
3
2
1
0
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 2030
In the following figure 4, we display the cumulative discovery, backdated and poorly adjusted
to a ”mean” value, with a 30-year translation, in order to facilitate the comparison with
cumulative production. The fit is not quite so good since 1960, suggesting that the MMS
model of reserve growth (the ultimate multiplier of 4.5 after 50 years) has become too high
due to the better initial appraisal with the help of modern technology.
Figure 4: US Lower 48 cumulative oil production and discovery shifted by 30 years
5
OPEC seminar
Jean Laherrere
Sept.28, 2001
US Lower 48: cumulative oil production & "mean"
discovery shifted 30 years
220
200
180
reserve growth
model too high?
160
140
120
100
80
60
mean discovery
disc. shifted 30 yr
production
40
20
0
1900
1920
1940
1960
1980
2000
2020
2040
-FSU (Former Soviet Union)
Since the FSU reserve data are maximum values (3P), one needs to reduce them to their
“mean” value to show a good fit between production and shifted discovery. A reduction of
45% allows such a good fit.
The following figure 5 provides a comparison between annual production and shifted annual
discoveries. It also suggests, simply by extrapolating the shifted discovery trend without any
sophisticated modeling, that production will, after a small rebound during the first half of the
present decade, decline significantly after.
Figure 5: FSU annual oil productions and annual shifted (by 20 years) discoveries (in green
the official values that correspond to the maximum i.e. 3P values, in blue the estimated mean
values)
6
OPEC seminar
Jean Laherrere
Sept.28, 2001
FSU: annual oil production & discovery smoothed on
5 years and shifted by 20 years and reduced by 45%
10
9
annual discovery
3P
8
annual discovery
reduced to "mean"
7
annual production
6
5
4
3
2
1
0
1930
1940
1950
1960
1970
1980
1990
2000
2010
2020
Figure 6: FSU cumulative oil productions and cumulated shifted (by 20 years) discoveries (in
green the official values that correspond to the maximum i.e. 3P values, in blue the estimated
mean values)
7
OPEC seminar
Jean Laherrere
Sept.28, 2001
FSU cumulative oil production & discovery (3P &
"mean") shifted by 20 years
350
discovery 3P
300
250
200
disc. 3P shifted
by 20 years
"mean" discovery
mean disc. shifted
by 20 years
production
150
100
50
0
1930
1940
1950
1960
1970
1980
1990
2000
2010
2020
As mentioned in the first part of this paper, this reduction of the official FSU figures is
consistent with the observations of Khalimov (1993). Evidence of over-estimation on reserves
is also clear in individual field production data. For instance, Samotlor, the largest oilfield, is
reported as having 28 Gb initial reserves whereas the extrapolation of the decline curve puts
ultimate recovery at 20 Gb as the following figure 7 shows.
Figure 7: Samotlor oil decline and estimate of ultimate recovery
8
OPEC seminar
Jean Laherrere
Sept.28, 2001
Samotlor oil decline
1300
1200
1100
1000
900
800
700
600
500
400
300
200
100
0
1964-1983
1983-1999
U scout = 3P
U decline = mean
0
4
8
12
16
20
24
28
-France
Exploration and production in France occurred in two cycles. The following figures 8 and 9
show the correlation between shifted discoveries and production. The annual plot (figure 8) is
of a “qualitative” nature. It shows how the troughs and peaks coincide, and even suggests that,
whereas the shift is 10 years for the first cycle, it might be 5 years only for the second. The
cumulative plot (figure 9) is more “quantitative”. It shows that there are too few fields for a
good fit: discoveries are made in quick succession but the time to deplete them is much
longer.
Figure 8: France annual oil production and discovery shifted by 10 years
9
OPEC seminar
Jean Laherrere
Sept.28, 2001
France: annual oil production & discovery shifted by
10 years
50
disc. smoothed on 7 years
& shifted by 10 years
production
40
30
20
10
0
1950
1960
1970
1980
1990
2000
2010
Figure 9: France cumulative oil production and discovery shifted by 10 years
10
OPEC seminar
Jean Laherrere
Sept.28, 2001
France: cumulative oil production and discovery
shifted by 10 years
1000
900
discovery
800
disc. shifted by 10 years
700
production
600
500
400
300
200
100
0
1920
1930 1940
1950
1960
1970
1980
1990 2000
2010
2020
-UK
The UK also displays two E&P cycles, with production peaks in 1985 and 1999. The rebound
of production in the 90s was predictable already in 1990 on the basis of the second cycle of
discovery (peak during the 80s). As for France, the correlation is more qualitative (see figure
10 based on annual values) than quantitative (see figure 11 based on cumulative values which
displays a poor fit on the details). The reason is that the large initial discoveries (85% of total
reserves) are concentrated over a few years whereas their production spreads into the second
cycle.
Figure 10: UK annual oil production and discovery shifted by 10 years
11
OPEC seminar
Jean Laherrere
Sept.28, 2001
UK annual production & annual discovery shifted by
10 years
2
1,8
discovery shifted 10 years
1,6
production
1,4
1,2
1
0,8
0,6
0,4
0,2
0
1970
1980
1990
2000
2010
Figure 11: UK cumulative oil production and discovery shifted by 10 years
12
OPEC seminar
Jean Laherrere
Sept.28, 2001
UK cumulative oil production & discovery shifted by
10 years
40
35
discoveries 2P
discovery shifted by 10 years
production
30
25
20
15
10
5
0
1950
1960
1970
1980
1990
2000
2010
2020
-World outside the Middle East (swing producers) & deepwater
The report “Energy for Tomorrow’s World: Acting Now!” published by the WEC in 2000
displays my correlation between discoveries and productions for the US Lower 48 and the
world outside the five large OPEC producers of the Middle East (these are the only countries
which significantly limit their production and capacity). But deepwater production (below 300
meters depth or 1000 feet) needs to be excluded as being a new cycle.
The fit between annual production and annual discovery is good with a 25-year shift. The raw
data on discovery is too high, and we have added the curve correcting the 3P values of FSU
into “mean” values. The fit suggests that production will peak soon and decline fairly strongly
for the next 20 years.
Figure 12: World outside “Middle East and deepwater”: annual oil production & discovery
shifted by 25 years
13
OPEC seminar
Jean Laherrere
Sept.28, 2001
World outside Middle East (swing producers) and
deepwater (>300 m): annual production and
discovery shifted by 25 years
35
30
discovery shifted by 25 years
discovery corrected FSU
production
25
20
15
10
5
0
1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 2030
The cumulative discovery plot shows a leveling towards an ultimate around 1200 Gb.
Cumulative production to 2000 is past the mid point (600 Gb) and will follow the corrected
shifted discovery for the next 20 years.
Figure 13: World outside” Middle East and deepwater”: cumulative oil production &
discovery shifted by 25 years
14
OPEC seminar
Jean Laherrere
Sept.28, 2001
World outside Middle East (swing producers) and
deepwater (>300 m): cumulative oil production and
discovery shifted by 25 years
1200
1000
discovery
disc. shifted by 25 years
discovery corrected on FSU
production
800
600
400
200
0
1880
1900
1920
1940
1960
1980
2000
2020
2040
-Middle East = swing producers
Since the oil shocks, Saudi Arabia, Kuwait, Iran, Iraq and the UAE have curtailed their
production deliberately, which therefore no longer fits the corresponding discovery trend. The
following figure 14 tentatively shows how unrestrained production might have evolved, based
on a 40-year shift.
Figure 14: Middle East annual crude oil production & discovery shifted by 40 years
15
OPEC seminar
Jean Laherrere
Sept.28, 2001
Middle East swing producers annual crude oil
production & smoothed discovery shifted by 40 years
25
disc. shifted by 40 years
production (OPEC data)
20
15
10
5
0
1950
1960
1970
1980
1990
2000
2010
2020
2030
2040
Cumulative production indicates that only 30% of the reported amount discovered have been
produced, assuming with some hesitation that “mean” values have been reported.
Figure 15: Middle East cumulative crude oil production & discovery shifted by 40 years
16
OPEC seminar
Jean Laherrere
Sept.28, 2001
Middle East swing producers: cumulative crude oil
production and discovery shifted by 40 years
800
700
discovery 2P
disc shifted 40 years
production (OPEC data)
600
500
400
300
200
100
0
1920
1940
1960
1980
2000
2020
2040
-Natural gas in North America
North America gas supply is important as being almost the only source, apart from LNG, to
meet demand. Electricity generation in California, which has already suffered from blackouts,
depends on gas supply.
The discovery (corrected to “mean”) shows a fairly good fit with production for a 20-year
shift giving a trough around 1985. However these data are still heterogeneous and unreliable,
suggesting that the governments need to make sure of obtaining more accurate data from the
operators (in particular on federal lands which are not supposed to be confidential). Most of
the recent production data are for dry production, being around 85% of the raw production.
The discovery trend relates only to conventional fields, so the comparison has to be with dry
minus unconventional production.
It seems that production will peak soon and decline strongly for the next decade, reflecting the
fall in discovery.
Figure 16: US +Canada +Mexico annual gas production & discovery shifted by 20 years
17
OPEC seminar
Jean Laherrere
Sept.28, 2001
US + Canada + Mexico natural gas: annual
production & discovery 2P shifted by 20 years
35
30
raw production
dry production
dry -unconventional.
25
disc. 2P shifted 20 yr
20
15
10
5
0
1920 1930 1940 1950 1960 1970 1980
1990 2000
2010 2020
Cumulative production displays a very good fit with cumulative shifted discovery from 1950
to the present. Production during the 90s seems to have drained the reserves faster, due to new
technology and infilled drilling.
Figure 17: US +Canada +Mexico cumulative gas production & discovery shifted by 20 years
18
OPEC seminar
Jean Laherrere
Sept.28, 2001
US+Canada+Mexico cumulative conventional
natural gas production & discovery shifted by 20
years
1600
1400
1200
discovery
disc. shifted 20 years
production raw-unconvent.
1000
800
600
400
200
0
1900
1920
1940
1960
1980
2000
2020
Cumulative gas discovery versus time shows a leveling, but it is much better to display its
trend by plotting it against the cumulative number of new field wildcats (“creaming curve”).
The creaming curve of figure 18 displays two hyperbolic curves, as do most of creaming
curves: the first one from pre 1900 to 1967 and a second from 1967 to the present (more
efficient exploration). With 1400 Tcf already discovered by about 400 000 new field wildcats,
the modeling shows that another 400 000 new field wildcats would bring only an additional
250 Tcf
Figure 18: US+Canada+Mexico creaming curve of gas discoveries versus number of new
field wildcats
19
OPEC seminar
Jean Laherrere
Sept.28, 2001
US+Canada+Mexico natural gas: creaming curve up
to 1997
1800
1600
1400
1200
H1 1900 U = 1400 Tcf
H2 1967 U = 500 Tcf
1000
800
H1+H2
cumulative discoveries
600
400
200
0
0
100
000
200
000
300
000
400
000
500
000
600
000
700
000
800
000
Official reports place great faith in future gas from the Gulf of Mexico (GOM). The MMS
publishes the details up to 1998 from 984 fields in the GOM OCS (Outer Continental Shelf).
It would be reasonable to assume that the data on discovery from this official source would be
of high quality, but a careful study shows many discrepancies. Some records even show
diminishing cumulative production, which is clearly impossible; and the comparison with
other confidential data sources shows large variation. However, there is a good correlation
between MMS discovery (grown to “mean” values) and production, showing that the GOM
production is about to peak and decline strongly. The fields in the deepwater area contain onethird gas (and two-thirds oil), presumably because of lower geothermal temperatures, whereas
those in the shallow waters contain two-thirds gas.
Figure 19: US GOM OCS annual gas production & discovery shifted by 20 years
20
OPEC seminar
Jean Laherrere
Sept.28, 2001
US GOM OCS annual gas "mean" discoveries smoothed
on 9 years, shifted by 20 years and correlated to annual
gas production from MMS database 1998
7
6
5
4
3
annual discovery: MMS
grown, smoothed 9 years,
shifted 20 years
annual production
2
1
0
1950
1960
1970
1980
1990
2000
2010
2020
-World’s deepwater
There is no consensus on the definition of the deepwater domain, which is variously placed at
between 200 m and 600 m; nor whether it should be treated as “conventional” or not. The
most common view, as adopted here, is to put the boundary at 300 m (>1000 ft) depth. Under
this definition, deepwater production started in the 80s. Albacora oilfield in Brazil, for
example, was producing in about 1000 m of water in 1987. So far, deepwater finds have been
made in only a limited number of basins, where few giants (>500 Mb) oilfields were
discovered in the 90s:1 giant (Crazy Horse) in the GOM; 4 in Brazil; 4 in Nigeria and 7 in
Angola. Annual deepwater production was about 0.5 Mb/d in 1994, rising to 3.5 Mb/d (1
Gb/a) in 2000. If discovery has now come close to peak, as seems likely, production might
peak at about 10Mb/d (3.5 Gb/a) around 2010-2015. Figure 20 shows annual production and
discovery curves with a 15-year shift, found from figure 21.
Figure 20: World’s deepwater (>300 m) annual oil production and discovery shifted by 15
years
21
OPEC seminar
Jean Laherrere
Sept.28, 2001
World's deepwater (>300 m) oil annual production &
discovery shifted by 15 years
7
discovery shifted by 15 years
6
production
5
4
3
2
1
0
1980
1990
2000
2010
2020
Cumulative discovery gives a fair correlation with production for a shift of 15 years. But as
the discovery curve has not reached maturity, we have plotted two logistic model with
ultimates of respectively 80 and 100 Gb, meaning that the undiscovered deepwater oil is
either 30 or 50 Gb compared to the 50 Gb already found.
Figure 21: World’s deepwater (>300 m) cumulative oil production and discovery shifted by
15 years with logistic modeling
22
OPEC seminar
Jean Laherrere
Sept.28, 2001
World's deepwater oil+condensate discovery and
production with the same logistic curve (shift 15
years) & two ultimates (80 & 100 Gb)
100
90
80
70
60
logistic disc U=100 Gb
logistic disc U=80 Gb
discovery
logistic prod U=100 Gb
logistic prod U=80 Gb
production
50
40
30
20
10
0
1970
1980
1990
2000
2010
2020
2030
-World
Economic and political constraints have meant that the 5 large OPEC producers of the Middle
East have produced below the capacity of their resource base. Accordingly, a good fit
between discovery and production is not to be expected. Further problems are introduced by
the fact that the discovery-production cycles occurred at different times in different parts of
the world; differences of geological habitat from dispersed, with many small fields, to
concentrated, with most reserves in a few very large fields; the nature of reservoirs, and the
differing economic conditions.
Nevertheless, such an approach does provide an aggregated view of reserves and production
in Gboe (= 6 Tcf). A shift of 37 years for oil, and 35 years for gas, provides a fair correlation.
It in turn allows some qualitative, if not quantitative, predictions for the next decades. Figure
22 provides such a comparison between the cumulative conventional oil and natural gas
discovery and production. The discovery data should be adjusted to a “mean” value, but that
is not yet possible.
Figure 22: World cumulative discovery and production for conventional oil and natural gas
23
OPEC seminar
Jean Laherrere
Sept.28, 2001
World: cumulative discovery and production of
conventional oil + condensate and natural gas
2000
1800
1600
1400
oil discovery
oil disc shift 37 yr
oil production
gas discovery
gas disc shift 35 yr
gas production
1200
1000
800
600
400
200
0
1900
1920
1940
1960
1980
2000
2020
2040
The preceding graph is for conventional oil and natural gas only. It has its limitations, given
the evidence of the growing contribution of unconventional hydrocarbons, and so it would
useful to present a simple model for the total production of all liquids and natural gas,
including the unconventional categories. In order to achieve such a model, it is necessary to
estimate the expected ultimate reserves and their depletion over time. This has been done in a
major 8-year study, made up of four large reports by Laherrere, Perrodon, Demaison and
Campbell. It gave the following ultimate reserve estimates (Perrodon et al, 1998): 2000 Gb of
conventional liquids, 750 Gb of unconventional liquids, 10 000 Tcf of conventional natural
gas and 2500 Tcf of unconventional gas. Figure 23 summarizes the conclusions with
production peaking around 2010 for all liquids and around 2030 for gas. Three years later, the
estimates for conventional hydrocarbons appear to be still valid on the basis of new data, but
the potential for unconventional hydrocarbons may have been exaggerated.
Figure 23: World’s conventional + non-conventional liquids and gas future production
24
OPEC seminar
Jean Laherrere
Sept.28, 2001
World liquids & gross gas future production from
ultimates (conventional + non-conventional) 2750 Gb and
12 500 Tcf (Perrodon, Laherrere, Campbell 1998)
30
past production
liquids
conventional + nonconventional liquids
25
conventional + nonconventional natural
gas 1 Gboe= 6Tcf
20
15
conventional oil
10
past production
gross gas
5
non-conventional oil
gas liquids
0
1925
1950
1975
2000
2025
2050
2075
2100
2125
It is interesting to note that these oil & gas production (for the world = consumption) profiles,
built from technical data, is completely out of the range of the IPCC 2000 scenarios (Third
assessment report) shown in figure 24. In my view (Laherrere 2001), these scenarios are pure
“wishful thinking” based on the unrealistic assumption of nearly limitless (and cheap) fossil
fuel resource (e.g. oceanic methane hydrates). One may wonder why the IPCC has chosen
such assumptions against the evidence?
Figure 24: World’s natural gas scenario from IPCC 2000 and from technical data
25
OPEC seminar
Jean Laherrere
Sept.28, 2001
IPCC 40 scenarios for gas and Laherrere's forecast
from technical data
1200
A1G message
1000
A1G AIM
A1 minicam
800
A1 AIM
technical data
600
data
400
200
0
1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
-Conclusion:
Most public reserve data are biased by politics, and are unreliable. They lead to false
conclusions, be they the optimistic view that there is no need to be concerned by a possible
scarcity of oil and natural gas, or the pessimistic view supporting extreme “green” attitudes,
according to which, the world is heading to a catastrophe in terms of climate change.
This paper shows that it is possible to work out the “mean” (expected) reserve values from the
technical data and then to correlate these “mean” annual and cumulative discoveries with
annual and cumulative production after a certain time shift, at least for the countries
producing without restraint. This shift allows us to forecast, qualitatively and, sometimes,
more precisely, what the future annual production will be.
The data described in this paper, indicates that the oil production outside OPEC Middle-East
would have started to decline in 1997 but for the arrival of deepwater oil and its fast rising
production. Excluding deepwater (the other unconventional oils do not influence the shape
because their production profile is a nearly linear gentle rise), non Middle-East crude oil
production has been level at around 46 Mb/d since 1988, apart from a “bump” in 1997 to 46.8
Mb/d, and will now start to decline. This overall decline is bound to continue and accelerate.
So the uncertainty is the extent to which deepwater production, as well as that from tarsands
and extra-heavy oil can compensate for the decline of the non Middle-East production.
These approaches need to be discussed and investigated in depth by the world community.
It is necessary to establish a politically neutral agency (or institute) to gather true “mean” data
on world’s reserves and productions.
26
OPEC seminar
Jean Laherrere
Sept.28, 2001
A step in this direction has already been made with establishment of the Association for the
Study of Peak Oil ("ASPO") and the Oil Depletion Analysis Centre (“ODAC”) in London,
which intend to gather the technical data, draw the conclusions and publicize the results.
A great contribution to stability would be made if the OPEC countries were to publish the true
values of production and reserves, and if the price were set at a level consistent with the
supply/demand balance. There would be less reason for concern that low prices would
damage the OPEC economies, and lessen the incentive for lower energy intensities. There
would also be less scope for unjustified extreme “greens” reactions. In the same way as oil
consumers need to understand the prime importance of energy for life, and the need to avoid
disruptions, oil producers have to realize that the best way to secure an overall agreement on
the fair price is to tell the truth about their reserves.
The developed and developing countries have to work together to determine what the reserves
are and to find out if the future production is as ominous as the shifted discoveries curves
suggest. They also need to a better understanding of the price setting mechanisms, based on
the cost of competing energies and on the impact of taxes in consuming countries. Finally,
they all need to maintain a fair and stable oil price of the benefit not only to the producers and
consumers of today, but also for our children and grandchildren.
OPEC’s goal of a fair price around 25$/b (in 2000 $) seems to be a good compromise.
Everyone has to come to understand the true implications of past discovery and production
trend, which demonstrate that the age of cheap oil has come to an end. OPEC also needs the
help and support of western governments in facing its extremely difficult task of managing
depletion.
-References
-Capen E.C. ,1996 "A consistent probabilistic approach"; SPE Reservoir Eng. Feb. 1/1
-Khalimov E.M., 1993 "Classification of oil reserves and resources in the Former Soviet
Union" AAPG 77/9 Sept. p.1636
-Khalimov E.M., M.V.Feign 1979 "The principles of classification and oil resources
estimation" WPC Bucharest, Heyden London 1980 p263-268
-Laherrère J.H. 2001 “Estimates of Oil Reserves ” IIASA International Energy Workshop
June 19-21 Laxenburg
http://www.iiasa.ac.at/Research/ECS/IEW2001/pdffiles/Papers/Laherrere-long.pdf
-Perrodon A., J.H. Laherrere, C.J.Campbell 1998 “The world’s non-conventional oil and gas”
Petroleum Economist March report 113p see http://textor.com/cms/dPEWNC.html
-USDOE/EIA-0534 1990 "US oil and gas reserves by year of field discovery» Aug.
-World Energy Council 2000 “Energy for tomorrow’s world –Acting Now“
27
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising