(Murray Duffin, July 2000)
Abstract
In this paper I will try to summarize & organize the salient information
from a great many source papers, to make the key points readily accessible.
First I will present conclusions, followed by definitions, data, commentary
on the data and arguments. Finally the key source papers are referenced,
with Internet addresses, for those who wish to dig deeper own their own. The
following is derived from two years and many hours of research.
1.0 Conclusions
The end of the era of cheap oil is at hand. It began in 1999 with the
resurgence of oil prices from a very uncharacteristic low, in essence a
bottom. We are not running out of oil, in fact far from it. However, we are
at the point where demand has caught up with supply, and given the huge
investment required, it will be 2- 3 years before slack supply is available
again, if ever. We are only a few years away from the time when 50% of all
of the originally recoverable conventional oil in the earth's crust will
have been exhausted. After that point (termed the Hubbert Peak), supply will
begin to decline, regardless of investment and the gap between supply and
demand will open.
OPEC is now, again, the world dominant supplier, and in a few years will
control more than 50% of the world's oil supply. Given OPEC control and
natural self interest, near term demand equal to or greater than supply,
medium term demand growing while supply diminishes and no readily available
alternatives, the price of oil is set to rise inexorably, possibly
punctuated by a few brief respites.
The next substantial price rise is likely to be as soon as Q4
2000. What will that do to the stock market and our economic boom? We must
adjust to (temporary?) shortages with high and increasing prices immediately
and to permanent shortages within a decade.
No one is to blame. A finite source will eventually get harder to produce
from and will yield at a lower rate. Less extravagant consumption could have
delayed, but not prevented this outcome. The economists who reassure us that
there is no foreseeable problem, and the politicians who listen to them have
got it wrong.
There is one other conclusion that I would like to note that is not directly
related to nor derived from this research. There is a great deal we can do
to reduce oil consumption, with only neutral or positive impact on the
economy. The know-how and technology already exist. Unfortunately, without
knowledge on the part of the people, and political will on the part of our
leaders, we are unlikely to act until prices and shortages really bite.
Dependence on oil could easily be cut by a factor of 4 per unit of GDP in
less than 20 years, without economic or environmental sacrifice.
2.0 Definitions (Terminology) and Explanations
Several of the definitions are accompanied by necessary elaborations, so
that the reader can understand the sources of disagreement and confusion
relative to the subject of petroleum supply.
2.1 First - some relationships
a) - Cumulative production (total from the first oil recovery to the
present)
b) - Reserves - oil that has been discovered, but not yet recovered
c) - Estimated Ultimately Recoverable (or Ultimate Recovery) EUR - how much
will have been produced when no further recovery is possible.
d) - Discovered-to-date - a) + b)
e) - Yet-to-find c) - d)
f) - Yet-to-produce c)-a) or d)+e)
2.2 Resources - Traditionally taken to be "total oil in the ground"
regardless of recoverability. Ivanhoe12 refers to it as "oil available with
someone else's money". Resources reflect the geologist's estimate based on
interpretation of exploration (usually seismic) data, prior to any drilling.
The geologist was motivated to "find" a lot of oil, in order to secure
financing. In early days (circa 70 years ago) resources could be as much as
10xEUR. Since the mid `70's with development of robust computer models,
extensive data bases, and digital 3D mapping, resource estimates are much
more accurate, about equal to EUR.
Ivanhoe notes that government agencies, academics and economists tend to
estimate resources (oil companies report reserves). They then assume that
"oil-in-the-ground" will someday be recoverable with improving technology,
when in fact much of it was never there in the first place. Be very wary of
any analysis or conclusions based on "resource" estimates.
2.3) Reserves - initial reserves are the estimate of oil available after
drilling. They usually reflect the engineer's conservative opinion of
recoverable in a known time, with a known technique and cost. Ivanhoe refers
to reserves as "oil available with our money".
With improved analytic techniques reserves are usually revised upwards.
Nearly all discoveries prior to the mid 70's have been back adjusted, so
reserves appear to grow, when in fact it is a result of refining originally
conservative estimates. Reserves can also "grow" because of reporting
regulations (SEC requires reporting of proven reserves), or for political
reasons - e.g. OPEC allocation of quotas based on reserves.
To add to the confusion there are 3 commonly used estimates:
Proven - i.e. 90% probability that EUR will be equal or higher
Proven + probable - i.e. 50% probability - the most likely case
Proven + probable + possible - i.e. 10% probability; very unlikely to be
realized.
Further, sometimes natural gas liquids (NGL) or non- conventional (also
termed unconventional) oil are reported as reserves.
The result is that summing reserves from all of the world's countries or oil
provinces, without critical analysis, will almost certainly lead to a too
high total. The detailed data needed to do the critical analysis resides in
closely held data bases (e.g. Petroconsult in Geneva) and is not available
to most economists, or even to the USGS Delphi participants. Economists and
the USGS therefore systematically overstate reserves.
It has been assumed that improved recovery techniques also cause reserve
growth. Evidence indicates that such new technology increases recovery rate
but has little or no impact on EUR. It does accelerate depletion. Current
reserves of course are c) - a) - e), EUR minus cumulative production, minus
yet-to-find.
2.4) Discoveries - New fields found through surveys and exploration.
Discoveries add to reserves. In a given "province" (e.g. - North Sea), the
large fields are the easiest to recognize from survey data, and are
therefore the first to be discovered. Subsequent discoveries tend to be
sequentially smaller.
2.5) Conventional Oil - Normally refineable crude oil, recoverable with
current or confidently planned technology and tools.
2.6) Non-conventional (Unconventional) Oil - Venezuela (Orinoco) bitumen,
Athabasca tar sands, shale oil and very very deep water offshore oil. Not
recoverable &/or refineable by conventional means
2.7) Economically recoverable - recoverable with a positive payback on
exploration + development + recovery costs, at a given price. Economically
recoverable may increase as price increases.
2.8) Energetically recoverable - recoverable with a net energy yield, when
burned as fuel, in excess of the energy used to build the equipment, supply
the material, drill the well and pump and transport the oil. A couple of
small North Sea fields have been exhausted only 2 years after being drilled.
It is possible (probable?) that they were economically but not energetically
recoverable, unless they were drilled with a used (energetically paid for)
oil platform.
2.9) Gb - One billion barrels or 109 barrels equals 1000Mb or one thousand
million barrels. Resources and reserves are normally expressed in Gb or Mb.
2.10) NGL - Natural Gas Liquids or gas condensate. Sometimes included in
reserves or EUR.
2.11) R/P - Reserves to Production Ratio - the number of years to total
exhaustion of reserves at the most recent year's production rate.
3) Data
3.1 EUR c) Variously estimated from as low as 500 Gb in the 1940's based on
poor data, to 3500 Gb in the 1960's based on anticipated impact of
technology that hasn't panned out. Estimates over the past 20 years have
converged on 2000 Gb20.
Woodward - review of 40 estimates - mean 2000 Gb, median near 2200Gb
Rand Corporation - between 1600 and 2000 Gb OPEC 2138 Gb
Petroconsult - 1800 Gb (plus 200 Gb NGL)
The author will use 2000Gb.
3.2 Cumulative Production a) Generally accepted as about 820 Gb at end
1999.
3.3 Yet-to-find e) Ranges from a low of about 150Gb for Campbell
(Petroconsult) to 600Gb for the USGS (United States Geological Survey). Both
the recently low and declining discovery rate, and Laherrere's parabolic 9
fractal analysis would support the lower number. The USGS number derives
from a Delphi exercise rather than analysis of hard data. The knowledge,
motivation of and influence on the Delphi participants are not known. The
author, being an optimist, has used 200Gb.
3.4 Reserves c) - e) - a)
While the reserves by country or province are subject to considerable
disagreement, the total tends to be accepted as near 1000Gb. Late 1995
published data will illustrator the range of disagreement - all in Gb17.
Region/Source OAJ WO Petro USGS
N. America 77 77 64 100
S. American 78 85 51 74
Europe 16 31 30 37
FSU (Former Soviet Union) 59 191 76 121
Africa 73 79 53 72
Middle East (ME) 660 590 439 583
Far East 42 51 38 62
Other 2 4 3 4
Total 1007 1108 754 1053
OGJ - Oil & Gas Journal, Petro - Petroconsult, WO - World Oil Journal,
USGS - US Geologic Survey
NOTES: . In 1987/88/89 ME reserves jumped by near 280 Gb without new
discoveries and with little new drilling. Some of the increase may have been
new analysis, but timing would suggest political motivation to influence
OPEC quota allocations.
ME-stated reserves have remained constant for several years, despite annual
production and lack of discovery.
FSU - has been strongly overstated in the past (10% probability numbers?).
Given the geography it could be expected to be substantially greater than
North America. Petroconsult discounts ME and FSU strongly (too much?) for
the above reasons.
The author goes with c) - e) -a) = 980 Gb.
3.5 Reserves Growth 17
Published reserves 1973 577 Gb
Published reserves 1996 1062 Gb
Production `73 - `96 513 Gb
Discoveries `73 - `96 520 Gb
Reserves "growth"`73 -`96 424 Gb
During 23 years reserves grew by nearly 40% of the sum of starting reserves
plus discoveries. However, it is likely that at least 130 Gb was political
(See 3.4) Real growth due to drilling and improved mapping and analysis may
have been 30% or about 13Gb/yr average. (It may have been much lower, also.
Notes: As new technology has now been applied to most old discoveries (Odell
17 notes a major effort on enhancing known fields during this period), and
as new discoveries were more accurate to start with, little further reserve
growth should be expected. Perhaps 3 to 4 Gb/yr will be realized during the
next decade.
3.6 Discovery
1945 - 1960 Averaged 35 Gb/yr - mainly due to ME
1970 - 1990 Averaged 23 Gb/yr - influenced by oil crises
1990 - 1999 Averaged 6 Gb/yr
NOTES: Discovery peaked in 1962/63 and the rate of decline since has been
accelerating. Campbell has noted that discoveries per dollar spent on
exploration was much lower in the `90's than in the `70's. The time from
discovery peak to production peak was 35 years in the lower
48,and 15 years in the FSU. World production is expected to peak 40-45 years
after the discovery peak. The American Association of Petroleum Geologists
has forecast a discovery rate of 8Gb/year from 2000 to 200919. They must
anticipate a huge increase in exploration which might happen given rising
oil prices.
3.7 Consumption in Gb/yr 17,23
1990 23.8 1995 24.9
1991 23.7 1996 25.4
1992 23.9 1997 26.1
1993 23.7 1998 26.2
1994 24.3 1999 26.7
2000 28.1 (projected)
NOTES . Discovery is now running less than 25% of consumption. Stock draw
down from Oct. `99 through Feb. 2000 was about 250 Mb6, or 4 months of the 3
recent OPEC increases. (16 months of the first increase)
3.8 What is a large discovery?
The largest field ever discovered was 90Gb - in Saudi Arabia. That is 3
years world supply in 2002 terms, assuming it is all recoverable. It isonly
1.5 years to the Hubbert Peak and declining recovery, if it was the sole
source.
- Today a big field would be 0.5 Gb
- The Caspian Sea referred to as a major find has an EUR of 30 Gb, enough to
postpone the Hubbert Peak about 6 months.
- 60 to 70% of known oil is in 300+ "Giant" fields
- 94% of known oil is in 1311 "Major" & "Giant" fields
- Of 17000+ (excluding 15000 termed insignificant) known fields worldwide,
the 2000 smallest contain a total of 50M bbl of oil - 6/10 of a day supply.
NOTES:
Of 17000 plus known significant fields, 7500+ are in North America. With
32000 fields worldwide (including insignificant) 20 it is unlikely that any
major finds have been overlooked. It is likely that no few of the smallest
2000, especially in deep water, have a negative net energy return.
3.9 When can we expect the Hubbert peak?
1999 Reserves 980 Gb
Possible reserve growth to 2020 70 Gb
Yet to find 200 Gb
Cumulative production 820 Gb
Initial resources 2070 Gb
50% depletion 1040 Gb
Already depleted 820 Gb
Remaining to Hubbert Peak 220 Gb
Average annual consumption 2000- 2010 - 31 Gb/yr
Hubbert Peak in 220/31 = 7 years - ie 2007
NOTES: - The author is a self-proclaimed optimist. For a much more
scientific analysis see Duncan and Youngquist 22 , who place the peak in
2006.
3.10 How rapid will the decline be after the peak?
The lower 48 production dropped by 50% in 37 years, about 1.8% per year.
North Sea fields that are already in decline are down 36% in 6 years for
Norway and 65% in 15 years for the UK, or about 6% per year 14. Campbell and
Laherrere project about 2.5 to 4% per year.
NOTES: If a demand growth of 1.5%/yr bumps into a supply decline of
3.5%/yr we will have a major problem. Energy efficiencies of greater than
2%/yr are probably achievable. Energy substitutions of 2% per year may be
possible. Allocation of the available supply will need to be addressed -
will market mechanisims serve?, which ones?. What if the decline is 6% per
year?
3.11) How about unconventional oil?
Unconventional "resource " estimates, vary from about 1300 -
4000 Gb (100 years' supply.? - No, only about 1/6 of it is considered to be
energetically recoverable.) However, the resources are meaningless for
purposes of this paper. What counts is the rate of recovery projected over
the next 20 years. In spite of nearly 30 years and several billion dollars
no successful technology has been developed to recover shale oil, and
efforts have largely been abandoned. Orinoco bitumen is already producing
(Orimulsion), and tar sands may be attractive at about $50.00 per barrel. It
is estimated that, with sufficient investment (billions of dollars) bitumen
and tar sand production could be brought to 2Mb/day in 2010 (about 2% of
needs) and 7 Mb/day in 2020. This addition would reduce the rate of decline
after the Hubbert Peak slightly.
Notes: One can imagine an energy efficient world in which North America
would need less than 10 Mb/day by 2050 and it could all come from tar sands,
at a high enough price.
3.12 What about year 2000? 16
Q1 Q2 Q3 Q4
Demand (Mb/day) 75.7 75.1 76.6 78.7
Supply non-OPEC 45.9 45.6 45.7 46.4
Supply OPEC 29.4 30.6 31.3 31.6
Supply TOTAL 75.3 76.2 77.0 78.0
NOTES: Excess supply in Q2 and Q3 does not make up prior destock. The swing
supplier is Iraq. They have no reason to help and have already announced
production problems. Total supply is above 98% of capacity in Q4, - no room
for problems.
3.13 What's wrong with R/P?
BP/Amoco statistics show that we have about 38.5 years of reserves at the
1998 production rate. This ratio is essentially a meaningless number as it
assumes implicitly that oil can be produced at a constant rate, until the
last drop is recovered, and then recovery will drop instantly to zero. The
real world doesn't work that way.
4.0 The Arguments
4.1 The Economists and Academics
The two perhaps most optimistic contributors are P. R. Odell and Michael
Lynch.
4.1.1 Odell 17
- Assumes that reserves growth in the future will be at least as high as in
1973-96, (very unlikely-see 3.5) but doesn't say how quickly. (It would have
to bequicker than in the reference 23 years to do much good.)
- Takes the USGS number for yet-to-be-discovered, the most optimistic
number,and ignores discovery rate.
- Assumes 35 Gb/yr production by 2020, but gives no detail on how we get
there from here.
- Doesn't discuss who controls the oil wells.
- Simply invents numbers for total non-conventional oil reserves with
noreference to credible sources & no consideration of costs or rate of
recovery.
- Assumes fairly explicitly that technology will take care of the details,
with no mention of what technology, or how quickly, or reference to past
progress. With this set of assumptions he concludes that there is no problem
in the foreseeable future.
4.1.2 Lynch 21
-In a 1998 paper Lynch tackles Campbell (the most profilic and often cited
geologist) head on. He notes that they both forecast the future in 1989 and
Lynch got it right. He points out that Campbell in 1989 and in subsequent
forecasts was consistently pessimistic, and has had to regularly raise his
reserve estimates and push out his day of reckoning.
-Bases his own arguments on the fact that reserves have always grown (up to
1998) (See 3.5) without questioning that they do this independent of
consumption and discoveries.
-Totally ignores the rapid decline in discovery rate, and seems to asume
that reserves will grow forever.
-States untapped resources as greater than 2000 Gb without citing sources.
He can only get to such a number by including the bulk of unconventional oil
and misusing "resources" (See 2.2) while ignoring issues of costs and rate
of recovery.
The not quite stated conclusion is that Campbell was wrong in the past and
must therefore be wrong in the future, while Lynch was right in the past and
must therefore be right in the future.
Lynch also concludes that there is no forseeable problem with the ongoing
availability of cheap oil.
His 1998 paper was published before prices tripled, so he may be a little
more cautious now.
4.2 The Geologists
The key references are Campbell, Laherrere, and Ivanhoe. They have all held
senior positions with major oil companies, they all have access to
considerable very specific data and they all understand the technology of
exploration, development and recovery. They seem to be unanimous on three
points:
- the world has been thoroughly surveyed and largely explored. There are no
major new surprises waiting to be discovered.
- big fields get discovered first. Anything not yet discovered is also not
big.
- the Hubbert Peak will be reached before 2010.
4.2.1 Campbell 1-5
-Is the most published, the most often cited, the most definite and probably
the most pessimistic; (or maybe realistic).
-Admits to being guilty as charged by Lynch, stating simply that as he gets
more or better data he revises his conclusions. The time frame may shift a
little, but the problem won't go away and won't be long postponed.
-Has probably been intentionally pessimistic, trying to get an uncaring
world to pay attention before it is too late, and has thus put himself in
the position of "crying wolf".
-Bases his argument mainly on discovery. If we are using oil 4-5 times
faster than we discover it, and if discovery per dollar of exploration is
declining rapidly, then we face a very finite limit on our supply.
-Along with Laherrere was the first to publish a Hubbert Peak analysis for
the world, and emphasizes that running out of oil is not the issue; it is
the declining production after the peak, when 50% of the oil is still in the
ground, that is the problem. It is the end of cheap oil that he warns us of.
-Also questions the declared reserves, noting: FSU reserves were exaggerated
under Communism for political reasons. FSU fields are in disarray, in
declining production, and won't help in the next years even if their
reserves are higher. ME reserves grew about 280 Bbbl from 1987-89 without
significant new drilling or exploration, probably because OPEC quotas were
pro-rated on reserves. ME reserves have not declined in the past 10 years
regardless of production and no new discoveries. Campbell declares reserves
at the low end of all estimates and thus positions the Hubbert Peak earlier.
4.2.2 Laherrere 7-11
-Is the most analytic and possibly the most interesting contributor. Of
particular interest is his analysis of the parabolic fractal distribution of
petroleum in the earth's crust. This analysis is probably the most telling
argument against major new discoveries.
-Also provides an excellent analysis of why the optimistic USGS numbers are
probably wrong.
4.2.3 Ivanhoe 12
-Provides some very useful clarification of terms and difficulties, and
otherwise largely agrees with Campbell.
4.3 Others
4.3.1 Riva 18
-Has major problems with arithmetic that leads him to some unsupportably
optimistic conclusions. Regretably he briefed Congress in 1998, and it is
most unlikely that any representatives caught the errors.
-Projects a demand of 94M bbl/day by 2010 or 34.8 B bbl/yeaar. Such a growth
rate would pull in the Hubbert Peak by about one year. It seems most
unlikely that such a high demand could be satisfied - so something has to
give.
4.3.2 Fleay 6
-Has published a recent (Feb. 2000) and very detailed paper, illuminating
the present supply/demand and price situation. This paper is a "must read".
4.3.3 Simmons 15,16
-Provides a comparison of 2000 vs 1973 and an analysis of 2000 with a
projection of the fourth quarter that is very alarming. Also a "must read".
5.0 What is to be done?
5.1 Immediately
- Start to accept with equanimity that higher gasoline and heating oil
prices are not going to go away, but indeed will continue to rise, and
adjust life styles accordingly.
-Stop looking for someone to blame.
-Stop looking for easy outs. As Peter Senge has noted "the easy way out
leadsback in".
-Communicate the facts to our fellow citizens and our political leaders.
5.2 Near term
-Declining oil availability poses a real threat to the economy, unless we
start preparing now. We can do 2 things:
-make the economy much more energy efficient;
-develop alternative (renewable) energy sources.
If we started now, in earnest, enough could be achieved before 2007 to avert
a crisis. Details of what can be done would require another paper.
REFERENCES
1) Campbell - Evolution of Oil asessments 3/23/2000
http://www.hubbertpeak.com/campbell/assessments.htm
2) Campbell - Myth of Spare Capacity 3/20/2000
http://www.hubbertpeak.com/campbell/mythcap.htm
3) Campbell - Letter to the Editor-Foreign Affairs 1/8/2000
http://www.hubbertpeak.com/campbell/foreignaffairs200001.htm
4) Campbell - The Imminent Peakof World Oil Production (Presentation to the
UK House of Commons 7/7/99
http://www.hubbertpeak.com/campbell/commons.htm
5) Campbell & Laherrere - The End of Cheap Oil - Scientific American v278/3
1996
http://www.dieoff.org/page140.htm
6) Fleay, Brian J - Oil Supply: The Crunch Has Arrived 3/13/2000
http://www.hubbertpeak.com/fleay/crunch.htm
7) Laherrere - Is USGS 2000 Assessment Reliable? 5/2/2000
http://www.hubbertpeak.com/laherrere/usgs2000/
8) Laherrere - Technological Progress or Bad Reporting and Bad
Arithmetic? - Geopolitics of Energy 22/4 4/16/99
http://www.dieoff.org/page176.htm
9) Laherrere - Parabolic Fractal Distributions in Nature French
Academy of Science 4/4/96
http://www.hubbertpeak.com/laherrere/fractal.htm
10) Laherrere -Multi-Hubbert Modeling - 7/97
http://www.hubbertpeak.com/laherrere/multihub.htm
11) Laherrere - Future Sources of Crude Oil Supply and Quality
Considerations 6/12/97
http://www.hubbertpeak.com/laherrere/supply.htm
12) Ivanhoe - Get Ready for Another Oil Shock - The Futurist Jan/Feb. `97
http://www.dieoff.org/page90.htm
13) Youngquist - Spending Our Great Inheritance - Then What? Geotimes 7/98
http://www.hubbertpeak.com/youngquist/geotimes.htm
14) Blanchard - The Impact of Declining Major North Sea Oil Fields upon
North Sea Production Jan 2000
http://www.hubbertpeak.com/blanchard
15) Simmons - The Oil World 1973 Compared to 2000 - Oil and Gas Journal
4/2000
http://www.simmonsco-intl.com/web/downloads/whitepaper.pdf
16) Simmons- The Energy Markets in 2000 - Running on Empty? Speech 5/24/00
http://www.simmonco-intl.com/web/downloads/whitepaper.pdf
17) Odell - A Guide to Oil Reserves and Resources
http://www.greenpeace.org/~climate/arctic99//reports/odell317.html
18) Riva - World Oil Production After Year 2000 - Business as Usual or
Crisis 8/18/95; Congressional Briefing
http://www.crie.org/n/e/eng-3.html
19) Shirley - Discoveries Are Getting Smaller - Explorer Jan 2000
http://www.aapg.org/explorer/archives/01_00/global_look.html
20) McKenzie, WRI - Estimated Ultimately Recoverable (EUR) Oil 3/96
http://www.wri.org/wri/climate/finitoil/eur-oil.html
http://dieoff.com/eur.htm
21)Lynch - Crying Wolf: Warnings About Oil Supply 3/98
http://www.stanford.edu/sep/jon/world-oil.dir/lynch/worldoil.html
22)Duncan and Youngquist - The World Petroleum Life Cycle 10/22/98
http://www.dieoff.com/page133.pdf
23)BPAmoco Energy Statistics
http://www.bpamoco.com/worldenergy/oil/index.htm
_______________________________________________
Crashlist resources: http://website.lineone.net/~resource_base
To change your options or unsubscribe go to:
http://lists.wwpublish.com/mailman/listinfo/crashlist
