ArticlePDF Available

Forecasting World Natural Gas Supply

Authors:

Abstract and Figures

World gas supply forecasting has proved difficult because its exploration, transportation, and customer bases depend so heavily on fluctuating economic factors. Our recent study showed that the conventional Hubbert model with one complete production cycle is not appropriate to use to forecast gas-production trends for most gas-producing countries. This paper presents our forecast for the world’s supply of conventional natural gas to Year 2050. We developed a “multicyclic Hubbert” approach that accurately models the gas-production history of each gas-producing country. Models for all countries were then used to forecast future production of natural gas worldwide. We present the multicyclic modeling approach in a convenient form that makes production data that exhibit two or more cycles easier to model and aggregated these models to regional and world levels. We also developed and analyzed supply models for some organizations [e.g., the Organization of Petroleum Exporting Countries (OPEC), the Organization for Economic Cooperation and Development (OECD), the E u ropean Union (EU), and the Intl. Energy Agency (IEA)]. Our results indicate that the world supply of natural gas will peak with a plateau production of 99 Tcf/yr from 2014 to 2017, followed by an annual depletion rate of 1%/yr. Regional analyses indicate that gas production of some regions will peak soon and that North American gas production is now (1999) at its peak. West European gas production is predicted to peak in 2002. Former Soviet Union (FSU) and Middle East countries, which contain appro x imately 60% of the world’s ultimate recoverable natural gas, will be the main sources of supply in the future .
Content may be subject to copyright.
Fo recasting World
Natural G as Supply
Fo recasting World
Natural G as Supply
S.M. Al-Fattah, SPE, and R.A. Startzman, SPE, Texas A&M U.
M a n a g e m e n t
6 2
M a n a g e m e n t
MAY 2000
Summary
World gas supply forecasting has proved difficult because its
exploration, transportation, and customer bases depend so
heavily on fluctuating economic factors. Our recent study
showed that the conventional Hubbert model with one com-
plete production cycle is not appropriate to use to fore c a s t
g a s - p roduction trends for most gas-producing countries.
This paper presents our forecast for the worlds supply of
conventional natural gas to Year 2050. We developed a
“multicyclic Hubbert approach that accurately models the
g a s - p r oduction history of each gas-producing country.
Models for all countries were then used to forecast future
p r oduction of natural gas worldwide. We present the mul-
ticyclic modeling approach in a convenient form that
makes production data that exhibit two or more cycles eas-
ier to model and aggregated these models to regional and
world levels. We also developed and analyzed supply mod-
els for some organizations [e.g., the Organization of Petro-
leum Exporting Countries (OPEC), the Organization for
Economic Cooperation and Development (OECD), the
E u ropean Union (EU), and the Intl. Energy Agency (IEA)].
Our results indicate that the world supply of natural gas
will peak with a plateau production of 99 Tcf/yr from 2014
to 2017, followed by an annual depletion rate of 1%/yr.
Regional analyses indicate that gas production of some
regions will peak soon and that North American gas pro-
duction is now (1999) at its peak. West European gas pro-
duction is predicted to peak in 2002. Former Soviet Union
(FSU) and Middle East countries, which contain appro x i-
mately 60% of the world’s ultimate recoverable natural gas,
will be the main sources of supply in the future .
Introduction
Natural gas is becoming an increasingly important source of
the world’s energ y. In recent years, natural gas use has gro w n
the fastest of all the fossil fuels, and it will continue to gro w
rapidly for several decades. The U.S. Energy Inform a t i o n
Admin. (EIA)
1
re p o r ted that world gas consumption gro w s
by 3.3%/yr compared with 2.2%/yr for oil and 2.1%/yr for
coal. This higher growth rate can be attributed to several fac-
tors. First, natural gas, including unconventional gas, is
available in abundant quantities in many parts of the world.
Second, natural gas is environmentally cleaner than coal and
c r ude oil. Third, the lower price of gas relative to other fuels
makes it attractive to many gas operators and consumers.
Fig. 1 shows the U.S. wellhead prices of gas and crude oil
since 1949. These data are wellhead inflation-adjusted
prices based on 1992 U.S. dollars on an equivalent-energ y
basis. The figure shows that a somewhat direct re l a t i o n s h i p
exists between oil and gas prices, with a time lag of 3 to 4
years. In 1949, the gas/oil price ratio was 0.12, indicating
that gas was 12% as valuable as oil on an energy basis. Since
that time, the trend of this ratio has been generally upward ,
reaching a value of 0.94 in 1998, indicating that gas has
now reached a close price parity with oil.
The gas industry is influenced by political events, eco-
nomic factors, and its relationship with the oil industry.
F i g . 2 shows the U.S. marketed-gas production rate since
1918. The gas-production trend from 1918 to 1970 shows
exponential growth. From 1970 to 1973, gas pro d u c t i o n
continued to increase but at a slower rate. Oil pro d u c t i o n
peaked in 1970. Contributing factors to the slowdown in
gas-rate increases might have been the oil-pro d u c t i o n
decline, which resulted in a decline of associated gas pro-
duction and lower gas prices. However, gas-supply short-
ages occurred during the very cold winter of 197273.
Actual gas production peaked in 1973, when OPEC cut
p r oduction of crude oil. Then, gas-production rates
d r opped, paralleling the decline in oil production. This
d r op in gas rate extended to 1975 because the gas market
was based on long-term gas-sales contracts with stable
prices. During 197579, gas production showed slow
g r owth and gas prices became more extensively re g u l a t e d .
In 1979, the Iranian revolution caused oil prices to incre a s e
s h a r p l y, reaching a peak in 1981. This corresponded to an
i n c r ease of gas prices, which peaked in 1984 (Fig. 1). The
oil/gas price time lag of 3 to 4 years possibly resulted fro m
the moderating effect of long-term gas contracts. In 1981,
with low gas demand, the “gas bubble (time period of high
gas re s e rves and production capacity and low demand) and
gas production decreased rapidly until 1986, despite the
fact that gas re s e rves and production capacities were high.
Since 1986, gas production increased steadily for a vari-
ety of reasons, including government policy and tax incen-
tives, increased gas demand caused by fuel switching and
low gas prices, and increases in unconventional gas pro-
d u c t i o n .
2
Of considerable interest to both producers and
consumers is the future direction of U.S gas pro d u c t i o n .
Our recent study
3
indicated that Hubbert ’s model,
4 - 7
w h i c h
p roved useful for oil-production forecasting, does not
account for fluctuations in gas-producing rates. Thus, it
may not be appropriate for forecasting gas production for
the U.S. and a number of other countries. This paper pre-
sents new forecasting models for the future gas supply. Our
supply models are based on country - b y - c o u n t ry pro d u c t i o n
analyses. We also discuss natural gas supply analysis by
region and by organization or group.
Copyright 2000 Society of Petroleum Engineers
This paper (SPE 62580) was revised for publication from paper 59798, originally presented
at the 2000 SPE/CERI Gas Technology Symposium held in Calgary. Original manuscript
received 14 January 2000. This paper has not been peer reviewed.
6 4
MAY 2000
Approach
Several authors
4 - 1 2
showed that Hubbert ’s model, with one
complete production cycle, is adequate for predicting cru d e
oil production. However, our recent study
3
showed that, in
the case of natural gas production, most countries exhibit
two or more Hubbert-type production cycles. These addi-
tional cycles apparently result from new exploration are a s
and technology, regulations, economic factors, and/or polit-
ical events. Using a Hubbert model with a single pro d u c t i o n
cycle does not allow for these factors. To account for addi-
tional production cycles, we used the multicyclic Hubbert
model described in Ref. 3. On the basis of the number of
cycles suggested by the production data, we can add a num-
ber of Hubbert-type production cycles. The production rate
of the multicyclic model can be computed with
, , . . . . . . . . . . .( 1 )
w h e re n=total number of production cycles, q
m a x
=m a x i-
mum or peak production, t
m a x
=time at peak pro d u c t i o n ,
and a=constant. Eq. 1 has fewer parameters (three) than
the Hubbert equation (which has four), making pro d u c-
tion data easier to model. Eq. 1 also provides a better fit for
multicyclic production data than the Hubbert model with
its single production cycle.
The parameters of the multicyclic model can be deter-
mined with a nonlinear least-squares method. Total ultimate
re c o v e ry, G
p a
or G
p a , u
, is then determined by adding the ulti-
mate recoveries for each production cycle.
. . . . . . . . . . . . . . . . . . . . . . . . . . .( 2 )
F u t u re recoverable gas is obtained by subtracting cumu-
lative production from ultimate re c o v e ry. The “logistic
c u r ve of the cumulative production of the multicyclic
model can be calculated as
. . . . . . . . . . .( 3 )
The multicyclic model proved to be an eff e c t i v e
a p p roach to modeling cyclic production data. The follow-
ing are some characteristics of the model.
It is derived from physical and mathematical concepts.
It can history match with good accuracy data fluctua-
tions influenced by economic factors and/or political
e v e n t s .
• The results are re p ro d u c i b l e .
• It uses obtainable historical data.
• These pro c e d u res are simple and can be readily imple-
mented with a computer spreadsheet pro g r a m .
The predictions can be updated easily with new data.
H o w e v e r, like Hubbert ’s original model, the results fro m
the multicyclic model are not unique because the model is
data sensitive, especially when few data from small fields
a r e used. There f o r e, we recommend perf o rming sensitivity
analyses of model parameters. Use of many data points
f r om large fields helps reduce model sensitivity.
Data Sources
We used historical natural gas production data from Refs.
13 through 17, U.S. gas-discovery data (1900–97) fro m
Refs. 17 and 18, and U.S. marketed-gas pro d u c t i o n
Fig. 1—U.S. gas and oil wellhead inflation-adjusted
prices and gas/oil price ratio, 1949–98 (all on equiva-
lent-Btu basis).
Fig. 3—Distribution of world’s conventional gas by
region: cumulative produced, future recoverable, and
ultimate recovery.
Fig. 2—U.S. natural gas and crude oil production (verti-
cal scale on equivalent-Btu basis).
North South and Western Eastern Middle Africa Asia-
America Central Europe Europe East Pacific
America and FSU
3,500
3,000
2,500
2,000
1,500
1,000
500
0
6 5
MAY 2000
6 5
(1918–97) from Refs. 15 and 16. Refs. 13 and 14 pro v i d e d
data for annual production of natural gas for all other
countries for 1971–97 and for proved re s e r ves of natural
gas for all countries.
Analysis by Region
This section presents analyses of world natural gas by
region. The forecasting model for each region was con-
s t r ucted by aggregating the corresponding countries mod-
els to their respective region level. Fig. 3 depicts the re s u l t s
for ultimate re c o v e r y, future recoverable gas, and cumula-
tive production for each re g i o n .
N o r th America. This region includes Canada, Mexico,
and the U.S. Historically, most of the re g i o n ’s production is
f r om the U.S. In 1971, U.S. production of 22.5 Tcf/yr con-
tributed approximately 88% of the re g i o n ’s pro d u c t i o n
c o m p a r ed with 72% in 1997. A 27-year (1971–97) average
regional production share for the U.S. is appro x i m a t e l y
81%. In contrast, Canada increased its share of pro d u c t i o n
f r om approximately 10% (2.5 Tcf/yr) in 1971 to 24% (6.6
Tcf/yr) in 1997. Mexico has a fairly stable share, ranging
f r om 3 to 4% of the total regional production during
1971–97.
Fig. 4 shows the re g i o n ’s actual and predicted gas pro-
duction from the multicyclic Hubbert approach. The 1997
p redicted production of 28.1 Tcf/yr is higher than actual
p r oduction by 0.5 Tc f / y r. The model indicates that con-
ventional gas production of this region peaks in 1999 at
a p p roximately 28.6 Tc f / y r. Production then declines at an
average annual rate of 0.7 Tcf/yr from 2001 to 2027; there-
a f t e r, the decline rate slows to 0.4 Tc f / y r. Most pro d u c t i o n
is predicted to come from the U.S. until 2010 when Cana-
dian and U.S. production is the same at 10.5 Tc f / y r. Cana-
d a ’s production share reaches approximately 55% by 2025
and decreases to 41% by 2040.
Our studies indicated that the estimated ultimate re c o v-
e r y of conventional gas for this region is appro x i m a t e l y
1,900 Tcf, with approximately 840 Tcf remaining to be pro-
duced (future recoverable gas) as of Ye a r-End 1997. This
quantity has an annual depletion rate of 3.3%/yr, the high-
est depletion rate of recoverable gas of any region world-
wide. In part i c u l a r, the U.S. has an annual depletion rate of
6 . 4 % / y r, ranking it worldwide as the country with the
highest rate. Canada and Mexico have annual depletion
rates of 2.0 and 0.6%/yr, re s p e c t i v e l y.
South and Central America. This region includes Arg e n t i-
na, Bolivia, Brazil, Chile, Colombia, Ecuador, Peru, Tr i n i d a d
and Tobago, and Venezuela. These countries pro d u c e d
about 3.4% of world gas production during 1997 and hold
a p p r oximately 4.4% of world proved re s e rves of natural gas
as of Ye a r-End 1997. The largest gas producer of this re g i o n
is Venezuela, followed by Argentina, with Trinidad and
Tobago coming in a far third. This same ranking holds for
p roved re s e rv e s .
Fig. 4 shows actual and predicted gas production for the
region. Our forecast model indicates that production in this
region reaches a plateau of slightly less than 5 Tcf/yr and
stays at this level from 2015 to 2021. Production then start s
to decline steadily at an average of 0.03 Tcf/yr through 2050,
re p r esenting an annual average decline rate of 0.7%/yr. The
model predicts production in this region at 3.8 Tcf/yr in
2050, with 285 Tcf of cumulative gas produced by that time.
Gas supply is mainly from Venezuela and Argentina until
the peak production of the region is reached, after which
Venezuela takes the lead. Venezuelan production is pre d i c t-
ed to contribute 54% of the total region production in 2025
and 78% in 2050.
Estimated ultimate re c o v e ry for this region is 419 Tc f ,
o f which we predict that 68% (285 Tcf cumulative gas)
w i l l be produced by 2050. Future recoverable gas of
t h i s region is 364 Tcf, which will last for appro x i m a t e l y
130 years, assuming a constant 1997 rate of pro d u c t i o n .
The re g i o n ’s annual depletion rate is 0.76%/yr; Arg e n t i n a
has the highest rate (2%/yr), followed by Trinidad and
Tobago (1.8%/yr).
We s t e rn Europe. The major producing countries fro m
this region will continue to be The Netherlands, the U.K.,
and Norw a y. These countries contributed 80% of 1997
west European production. Norw a y ’s share of the re g i o n ’s
p r oduction is predicted to increase from 15.4% in 1997 to
51% in 2015, after which it dominates the re g i o n ’s pro-
duction, reaching a share of 85% in 2035 and 90% in 2050.
Fig. 5 shows the re g i o n ’s actual and predicted pro d u c-
tion. Our model shows that peak production occurs in
2002 at 12 Tc f / y r. Production then has an annual average
Fig. 4—Western Hemisphere region natural gas produc-
tion models.
Fig. 5—European region natural gas production models.
North America
South and
Central America
Model
Eastern
Europe
Western
Europe
Model
MAY 2000
decline rate of 3.6%/yr until 2015. After that, pro d u c t i o n
gets flatter (mainly owing to Norw a y ’s production), then
has a more relaxed annual decline rate of 2%/yr until 2050.
Some countries in this region have already passed their
p roduction peaks. Austria, which had a peak in 1975 and
a second smaller peak predicted to have occurred in 1998,
is now in decline. France peaked in 1983 and has experi-
enced a general decline since then.
We estimated the ultimate recoverable gas for this re g i o n
to be 560 Tcf. Of this amount, approximately 66%
remained to be produced as of Ye a r-End 1997 and 535 Tc f
of cumulative gas is predicted to be produced by 2050.
Our analysis indicates that the west European re g i o n ,
w i t h an annual depletion rate of 2.8%/yr, has the second-
highest depletion rate after North America. Countries with
high depletion rates are Denmark (6.3%/yr), France
(5.9%/yr), Austria (5.5%/yr), and The Netherlands
(5.5%/yr). With 1997 production given as constant, future
re c o v e ry of gas will continue for 36 years, close to that of
N o r th America.
E a s t e rn Europe and FSU. This region includes Albania,
H u n g a ry, Romania, other low-producing countries of east-
e r n Europe, and the FSU. As of 1997, the FSU alone
accounted for 29% of the worlds gas production and held
a p p roximately 39% of the worlds proved gas re s e rves. In
contrast, eastern European nations contribute less than 2%
of world gas production and have less than 1% of the
w o r l d ’s proved re s e rv e s .
Fig. 5 shows actual and predicted production of this
region. Predicted 1997 production is slightly higher than
the re p o rted production by 0.6 Tc f / y r. Production in this
region has been declining since 1990, when the countries
of FSU separated and became independent. Production is
p redicted to stay at about the 1997 level until 1999 and to
s t a r t increasing by 2000, reaching a peak in 2032 at
a p p roximately 36 Tc f / y r. Our prediction indicates a pro-
duction plateau at the peak level extending from 2030 to
2035. With the exception of FSU, all eastern Euro p e a n
countries passed their production peaks in the 1980s. The
i n c r eased gas production is expected to come mainly fro m
the FSU, primarily from western Siberia. The FSU’s low
gas-depletion rate of 0.9%/yr is far less than eastern
E u ro p e ’s high depletion rate, which averages appro x i m a t e -
ly 5%/yr. Estimated ultimate re c o v e r y for this region is
3,411 Tcf, with approximately 81% of this amount re m a i n-
ing to be produced as of Ye a r-End 1997. By 2050, about
72% of the ultimate re c o v e ry is predicted to be pro d u c e d
and gas production is predicted to be 30.6 Tc f / y r, appro x i-
mately the 1990 level.
Middle East. This region includes Bahrain, Iran, Iraq,
Kuwait, Oman, Qatar, Saudi Arabia, Syria, and the UAE
among other countries. It contains 34% of the worlds
p r oved gas re s e rves. The major gas-producing countries
in this region are Saudi Arabia and Iran, followed by the
UAE and Qatar. Political events in the region and oil-
related OPEC policies affected the re g i o n ’s gas pro d u c-
tion. These include Iran’s cutback production of oil in
1979, the United Nations (UN) sanction on Libya, and
the UN sanction on Iraq following its invasion of Kuwait
in 1990. Fig. 6 shows the re g i o n ’s actual and pre d i c t e d
p r oduction, indicating how effective the multicyclic
a p p r oach is in accounting for such unpredictable events.
P r oduction is predicted to increase at an average annual
i n c r ease of 0.6 Tcf/yr until 2040, when it peaks at 29.3
Tc f / y r. Production then declines at a low rate of appro x i-
mately 0.2 Tc f / y r. By 2050, production is predicted to be
27.2 Tc f / y r, with most of it from Iran, Saudi Arabia, Qatar,
and the UAE.
Estimated ultimate re c o v e ry for the Middle East is 2,508 Tc f ,
with future recoverable gas of 2,433 Tcf remaining as of Ye a r -
End 1997. Our work indicates that approximately 43% of this
re g i o n ’s future recoverable gas will be produced by 2050.
Africa. The 1997 production for this region was 3.3 Tc f / y r,
of which Algeria produced approximately 74%. Algeria is
expected to continue to be the major producer in Africa
until 2043, when Nigeria and Libya take over the role. By
2050, Nigerian, Libyan, and Algerian production share s
a re predicted to be 41, 29, and 17%, re s p e c t i v e l y. Fig. 7
shows actual and predicted production of the re g i o n . T h e
f i g u re indicates that production peaks in 2014 at 7.2 Tc f / y r,
with an average annual increase of 0.24 Tc f / y r. Pro d u c t i o n
then declines at an average annual of 0.17 Tcf/yr until 2043
and gets flatter thereafter with less than a 0.05-Tcf/yr aver-
age annual decline. The model shows that the 1997 pro-
duction level is repeated in 2035.
Fi g. 6Middle East region natural gas production model.
Fig. 7—Asia Pacific and Africa regions natural gas pro-
duction models.
6 6
6 9
MAY 2000
6 9
Estimated ultimate re c o v e ry for this region is 476 Tc f ,
and the future recoverable gas is 426 Tcf (Table 1 of Ref.
19). At 1997 production levels, remaining recoverable gas
is expected to last more than 10 decades. This re g i o n
model gives an annual depletion rate of 0.78%/yr, close to
that of the South and Central America re g i o n .
Asia Pacific. P roduction from this region has incre a s e d
steadily since 1971, except during 1982 and 1983 when
Indonesia reduced its production to 44 and 20%, re s p e c-
t i v e l y. This had an impact despite the addition of Malaysia
and Thailand’s production as first re p o rted in 1981 and
1982, re s p e c t i v e l y. Fig. 7 shows the re g i o n ’s actual and pre-
dicted production, indicating that peak production is pre-
dicted to occur in 2012 at 16.7 Tc f / y r. Production decline
then takes place at an annual average of approximately 0.4
Tcf/yr (2.4%/yr). We estimate ultimate re c o v e r y of gas for
this region to be about 800 Tcf, of which 86% remained to
be produced as of Ye a r-End 1997 at an annual depletion
rate of 1.2%/yr.
Analysis by Organization/Group
We combined the country models according to their aff i l-
iated organization or group. This section presents analyses
of natural gas production and outlook of gas supply for
OPEC, OECD, IEA, and the EU. Figs. 8 and 9 show the
results for each organization or group. Ref. 19 gives the
p rediction-model plots for each of these org a n i z a t i o n s .
OPEC and non-OPEC. The OPEC countries hold appro x-
imately 43% (2,175 Tcf) of the world’s proved re s e rves of
natural gas as of January 1998. Iran, Qatar, the UAE, and
Saudi Arabia contribute 69% of OPEC gas proved re s e rv e s
and approximately 30% of worldwide re s e rves. Although
gas production from OPEC countries accounts for only
13% of 1997 world produced gas, its production has
i n c reased steadily since the 1970’s. Gas pro d u c t i o n
i n c r eased from 2 Tcf/yr in 1971 to 10.5 Tcf/yr in 1997, or
81%. Exceptions to this continual production incre a s e
o c c u r red in 1974 and in 1979–80, when gas pro d u c t i o n
d e c reased with the OPEC cutback in production of cru d e
oil in 1973 and Iran’s revolution in 1 9 7 9 .
Fig. 9 shows actual and predicted gas production for
OPEC and non-OPEC countries. The figure indicates that
gas production from OPEC countries peaks in 2038 at 34.9
Tc f / y r, with most production coming from Iran, Saudi Ara-
bia, Qatar, and the UAE through the forecast period. In
contrast, conventional gas production from non-OPEC
countries is expected to peak by 2004 at 78.8 Tc f / y r, then
d e c r ease at an average decline of 1.1%/yr of peak pro d u c -
tion to 2050.
O P E C ’s future recoverable gas is estimated to be 3,055
Tcf as of Ye a r -End 1997, with an annual depletion rate of
less than 0.5%/yr. These future re s e rves account for
a p p roximately 39% of the worlds future recoverable gas.
OECD and non-OECD. Fig. 9 shows actual and pre d i c t-
ed OECD and non-OECD gas production. The model indi-
cates that the OECD conventional gas production peaks in
2001 at 42.2 Tc f / y r. Most of this org a n i z a t i o n ’s pro d u c t i o n
comes from North American countries: the U.S. and Cana-
da. The annual OECD gas-depletion rate is appro x i m a t e l y
3 % / y r, and ultimate recoverable gas is 2,564 Tcf, with
1,300 Tcf recoverable gas remaining to be produced as of
Ye a r-End 1997.
Non-OECD production is tightly controlled by the high
p r oduction of the FSU. Peak production is expected to
occur in 2031 at 77.4 Tc f / y r, with a relatively low gas-
depletion rate of 0.65%/yr. This gro u p ’s ultimate re c o v e r-
able gas is 7,480 Tcf, re p resenting approximately 83.5% of
the world’s future recoverable gas. Most future gas pro d u c-
tion from this group will come from the FSU and the Mid-
dle East gulf countries, including Iran.
EU and IEA. EU gas production accounted for only 10%
of the worlds produced gas in 1997, and the area holds
only approximately 2% of the world’s proved re s e rves. Fig.
9, which shows actual and predicted EU gas pro d u c t i o n ,
indicates that it peaks as early as 2001 at 10 Tc f / y r. Pro-
duction then declines to an insignificant amount by 2050.
Among organizations and groups considered in this study,
EU has the highest annual gas-depletion rate at 4.4%/yr.
Ultimate gas re c o v e r y obtained from the EU model is 366
Tcf, with 193 Tcf future recoverable gas remaining as of
Ye a r- E n d 1 9 9 7 .
Fig. 9 also shows the IEA gas production and fore c a s t
model. The 1997 production increases until 2001, when it
peaks at 40 Tc f / y r. Production then declines steadily at an
Fig. 8—Distribution of world’s conventional gas by
organization/group.
Fig. 9—Natural gas production predictions for different
organizations/groups and the world.
MAY 2000
annual depletion rate of 3.4%/yr. Ultimate recoverable gas
obtained from the IEA model is 2,310 Tcf, of which
a p p roximately 1,100 Tcf of future gas remained to be pro-
duced as of Ye a r-End 1997.
The World
World marketed-gas production increased from appro x i-
mately 41 Tcf/yr in 1971 to approximately 82 Tcf/yr in
1997, a 100% increase over the 27-year period at a pro-
duction growth rate of about 4%/yr. Fig. 9 shows our
world conventional natural gas prediction model and indi-
cates a very good match with the fluctuating historical pro-
duction data. Predicted world production of gas in 1997
(83.8 Tcf/yr) is higher than actual production by 1.5 Tc f .
The model indicates that world production of natural gas
peaks between 2014 and 2017 at an approximately flat rate
of 99 Tc f / y r . After the peak is passed, production starts to
decline gradually and the curve gets flatter.
Between 1967 and 1999, the worlds proved natural gas
re s e rves increased substantially from about 1,043 to 5,145
Tcf, an average annual increase of 128 Tcf. Appro x i m a t e l y
72% of the 1999 worlds proved re s e r ves is in the FSU and t h e
Middle East region. Our results indicate that world ultimate
re c o v e ry of conventional gas is 10,000 Tcf, of which future
recoverable gas is approximately 7,900 Tcf as of Ye a r- E n d
1997. Tables 1 and 2 of Ref. 19 provide our results for world
gas grouped by region and organization, re s p e c t i v e l y.
Reserves/Production Ratio and
Annual Depletion Rate
The worlds gas re s e rv e s / p roduction (R/P) ratio is 96 years,
which means that world gas re s e rves will last for 96 years
if the 1997 rate of production is constant in the future. Use
of the R/P ratio as an indication of future production is
misleading and meaningless because production rates
p robably do not remain constant over a long period of time
and then drop suddenly to zero when re s e rves are deplet-
ed. If it is needed, an R/P curve as a function of time (or as
a function of production rate) can be constructed with pre-
dicted production rate with a given estimated ultimate
re c o v e ry, G
p a , u
, value (Fig. 10). As an example, at Ye a r
2030, the world’s remaining gas re s e rves would last for
a p p r oximately 50 years provided the predicted rate of pro-
duction in 2030 remained unchanged.
A better alternative is the annual depletion rate, which is
annual production divided by remaining re s e rv e s
e x p ressed in percentage. It is a measure of how fast the
re s e rves are being depleted each year at that years rate of
p roduction. Annual world-gas-depletion rate for 1997 is
computed at 1%/yr. Fig. 10, which shows world-gas-deple-
tion rate determined with the predicted rate of pro d u c t i o n
and the obtained value of future recoverable gas, indicates
it increases throughout the forecast period and re a c h e s
a p p r oximately 2.3%/yr by 2050.
Conclusions
We presented our analyses of the future of the worlds con-
ventional natural gas by region and org a n i z a t i o n / g ro u p .
P roduction data from several gas-producing countries or
regions showed fluctuations. These data were affected by
the relationship between the gas and oil industries, eco-
nomic burdens, and governmental-policy implementa-
tions. The multicyclic model was an effective approach for
modeling such production trends and developing fore c a s t-
ing models for them.
Our analyses indicate that most industrialized countries
(e.g., the U.S., Denmark, France, and the U.K.) are depleting
their gas re s o u rces much faster than are developing countries.
Fuel switching and gas dependence by industrial and com-
m e rcial sectors and production decline of crude oil in these
countries are among the reasons for the high depletion rate.
This means that gas production of some regions is now in
decline or will peak soon. North American gas production is
p r edicted to have peaked in 1999 at a rate of approximately 29
Tc f / y r, and western European gas production is expected to
p e a k in 2002 at 12 Tc f / y r. However, the FSU and the
m a j o r Middle East gulf countries (Iran, Saudi Arabia, Qatar,
and the UAE), which hold 68.5% of world proved re s e r ves of
natural gas, will be major sources of world gas supply with
4,880 Tcf of future recoverable gas, re p resenting appro x i m a t e -
ly 62% of the worlds future re c o v e r y of natural gas.
Nomenclature
a= constant, 1/t, 1/yr
G
p a
= ultimate re c o v e ry of gas, L
3
, Tc f
G
p a , u
= estimated ultimate re c o v e ry, L
3
, Tc f
n= number of production cycles
q
m a x
= peak production rate, L
3
/t, Tc f / y r
q(t)= p roduction rate as a function of time, L
3
/t, Tc f / y r
Q= cumulative production, L
3
, Tc f
t= time, t, calendar year
t
m a x
= time at peak production, t, calendar year
Acknowledgment
S.M. Al-Fattah thanks Saudi Aramco for supporting his
PhD study at Texas A&M U.
References
1 .“ I n t e rnational Energy Outlook 1998, DOE/EIA-0484, Off i c e
of Integrated Analysis and Forecasting, U.S. Dept. of Energ y,
EIA, Washington, DC (April 1998).
2 . Wa t t e n b a rg e r, R.A. and Villegas, M.E.: “Trends in U.S. Natur-
al Gas Production,” Advances in the Economics of Energy and
R e s o u rc e s, J.R. Moroney (ed.), J.A.I. Press, Greenwich, Con-
necticut (1995) 9 , 1 6 9 .
3 . Al-Fattah, S.M. and Startzman, R.A.: “Analysis of Wo r l d w i d e
Natural Gas Production,” paper SPE 57463 presented at the
Fig. 10—World historical and predicted annual deple-
tion and R/P ratio of conventional gas.
J P T
J P T
7 0
7 2
MAY 2000
1999 SPE Eastern Regional Meeting, Charleston, West Vi r -
ginia, 20–22 October.
4 . H u b b e r t, M.K.: “Nuclear Energy and Fossil Fuels,Drill. &
P rod. Prac. (1956) 17.
5 . H u b b e r t, M.K.: Energy Resources,” Publication 1000-D,
Natl. Academy of Science/Natl. Research Council (1962).
6 .H u b b e rt, M.K.: “Degree of Petroleum Exploration in the
United States,” AAPG Bull. (11 November 1967) 5 1 , 2 2 0 7 .
7 . H u b b e r t, M.K.: “Techniques of Prediction as Applied to Pro-
duction of Oil and Gas, P ro c ., U.S. Dept. of Commerc e
Symposium, Washington, DC (June 1980) 16.
8 . Ivanhoe, L.F.: “Future World Oil Supplies: There is a Finite
Limit, World Oil (October 1995) 77.
9 . Campbell, C.J.: The Coming Oil Crisis, Multi-Science Publish-
ing Co. and Petroconsultants S.A., Brentwood, U.K. (1997)
8 6 .
1 0 . A l - J a rri, A.S. and Startzman, R.A.: “Worldwide Petro l e u m -
Liquid Supply and Demand,” J P T (December 1997) 1329.
1 1 . Campbell, C.J. and Laherre re, J.H.:The End of Cheap Oil,
Scientific American ( M a rch 1998) 78.
1 2 .B a rtlett, A.A.: “An Analysis of U.S. and World Oil Pro d u c-
tion Patterns Using Hubbert-Style Curves,” M a t h e m a t i c a l
G e o l o g y (August 1999) 11.
1 3 . E n e rgy Statistics Sourc e b o o k, 13th edition, OGJ e n e rgy data-
base, PennWell Publishing Co., Tulsa, Oklahoma (1998).
1 4 . I n t e rnational Energy Statistics Sourc e b o o k, eighth edition,
OGJ e n e rgy database, PennWell Publishing Co., Tu l s a ,
Oklahoma (1998).
1 5 . Twentieth Century Petroleum Statistics, 52nd edition, DeGol-
yer and MacNaughton, Dallas (1996).
1 6 . Twentieth Century Petroleum Statistics, 54th edition, DeGol-
yer and MacNaughton, Dallas (1998).
1 7 . EIA, Internet Home Page: http://www. e i a . d o e . g o v / .
1 8 . Attanasi, E.D. and Root, D.H.: “The Enigma of Oil and Gas
Field Growth,AAPG Bull. ( M a r ch 1994) 7 8 , 3 2 1 .
1 9 . Al-Fattah, S.M. and Startzman, R.A.: “Forecasting Wo r l d
Natural Gas Supply, paper SPE 59798 presented at the
2000 SPE/CERI Gas Technology Symposium, Calgary,
3 – 5 A p r i l .
SI Metric Conversion Factors
b b l ×1.589 873 E0 1 = m
3
B t u ×1.055 056 E+0 0 = k J
f t
3
×2.831 685 E0 2 = m
3
Saud M. Al-Fattah is a PhD-degree candidate in the Pe t r o-
leum Engineering Dept. at Texas A&M U., College Station,
Texas, on an educational leave of absence from Saudi Arabi-
an Oil Co. (Saudi Aramco). With Saudi Aramco since 1985,
he has worked as a reservoir engineer in the Abqaiq Re s e r-
voir Management Div. and as a reservoir simulation systems
analyst in the Petroleum Engineering Applications Services
Dept. His specialties include reservoir engineering, opera-
tions research, economics evaluation, and strategic planning.
A l - Fattah holds BS and MS degrees in petroleum engineering
from King Fahd U. of Petroleum and Minerals. Richard A.
S t a r t z m a n is a professor in the Petroleum Engineering Dept.
of Texas A&M U. in College Station. He holds a BS degree
from Marietta College and MS and PhD degrees in petrole-
um engineering from Texas A&M U. A Distinguished Member,
Startzman has published widely in the petroleum literature.
... Natural gas plays a very important role in economic life, it has increased in recent years the use of natural gas, which continues to grow rapidly for several decades [1]. ...
... Copyright © 2020 MECS I.J. Mathematical Sciences and Computing, 2020, 5, 37-46 Table 5. Akaike information criterion Model AIC ARIMA (5,1,5) 2.41 ARIMA (9,1,9) 2.36 ARIMA (9,1,5) 2.36 ARIMA (5,1,9) 2.35 ...
... Copyright © 2020 MECS I.J. Mathematical Sciences and Computing, 2020, 5, 37-46 Table 5. Akaike information criterion Model AIC ARIMA (5,1,5) 2.41 ARIMA (9,1,9) 2.36 ARIMA (9,1,5) 2.36 ARIMA (5,1,9) 2.35 ...
... Of these models, the Hubbert model is the most widely used, which was used by 8 out of the 13 studies analysed (i.e. Al-Fattah and Startzman, 2000;Laherrere, 2002;Imam et al., 2004;Zerta et al., 2008;Nel and Cooper, 2009;Valero, 2010, 2011;Maggio and Cacciola, 2012). Actually, this is not just the case for natural gas production forecasts. ...
... Bell-shaped curve-fitting models can in turn be divided into three categories, i.e. symmetrical curve models, positivelyskewed curve models and negatively-skewed curve models . Currently, symmetrical curve models (the well-known Hubbert model is one example) are the most widely used in the literature (Al-Fattah and Startzman, 2000;Laherrere, 2002;Imam et al., 2004;Zerta et al., 2008;Nel and Cooper, 2009;Valero, 2010, 2011;Maggio and Cacciola, 2012). However, due to the problematic assumption of symmetrical curve shape versus actual observation on post-peak regions, positively-skewed curve models are recommended and used here to forecast long-term natural gas production. ...
... For this reason, it is important to determine the number of production cycles to use. Current multi-cycle modelling usually chooses the number of production cycles according to the researchers' subjective view or qualitative analyses of historical production data (see examples in Al-Fattah and Startzman, 2000;Imam et al., 2004), which can make the results highly uncertain since different researchers may use different number of cycles even though the same historical production curve is used. Therefore, establishing a quantitative method to determine the number of production cycles is required . ...
Article
Full-text available
As the cleanest fossil fuel in terms of carbon dioxide (CO2) emissions, natural gas demand is expected to increase rapidly in future due to its important role in the transition of the world energy system. In this case, understanding potential limits to future production of the world’s natural gas resources becomes increasingly important. This paper uses a modified multi-cycle generalized Weng model to forecast the long-term production of natural gas by region, and also globally. Both conventional and unconventional gas production are considered. Our results show that world natural gas production is likely to peak in the range 3.7 to 6.1 trillion cubic meters per year (tcm/y) between 2019 and 2060 depending on assumptions made on the size of the global ultimately recoverable resource (URR) of natural gas. A comparison of this paper’s forecasts with those from the scientific literature and from major energy institutes shows that the projection in this paper’s ‘high scenario’ can be seen as a likely upper-bound on future global natural gas production. To turn this upper-bound projection into reality, great efforts will be needed from the gas industry to discover more conventional and unconventional gas resources, and to make these recoverable. Full paper: https://www.sciencedirect.com/science/article/pii/S2352484720304200
... Of these models, the Hubbert model is the most widely used, which was used by 8 out of the 13 studies analysed (i.e. Al-Fattah and Startzman, 2000;Laherrere, 2002;Imam et al., 2004;Zerta et al., 2008;Nel and Cooper, 2009;Valero, 2010, 2011;Maggio and Cacciola, 2012). Actually, this is not just the case for natural gas production forecasts. ...
... Bell-shaped curve-fitting models can in turn be divided into three categories, i.e. symmetrical curve models, positivelyskewed curve models and negatively-skewed curve models . Currently, symmetrical curve models (the well-known Hubbert model is one example) are the most widely used in the literature (Al-Fattah and Startzman, 2000;Laherrere, 2002;Imam et al., 2004;Zerta et al., 2008;Nel and Cooper, 2009;Valero, 2010, 2011;Maggio and Cacciola, 2012). However, due to the problematic assumption of symmetrical curve shape versus actual observation on post-peak regions, positively-skewed curve models are recommended and used here to forecast long-term natural gas production. ...
... For this reason, it is important to determine the number of production cycles to use. Current multi-cycle modelling usually chooses the number of production cycles according to the researchers' subjective view or qualitative analyses of historical production data (see examples in Al-Fattah and Startzman, 2000;Imam et al., 2004), which can make the results highly uncertain since different researchers may use different number of cycles even though the same historical production curve is used. Therefore, establishing a quantitative method to determine the number of production cycles is required . ...
Article
Full-text available
Abstract As the cleanest fossil fuel in terms of carbon dioxide (CO2) emissions, natural gas demand is expected to increase rapidly in future due to its important role in the transition of the world energy system. In this case, understanding potential limits to future production of the world’s natural gas resources becomes increasingly important. This paper uses a modified multi-cycle generalized Weng model to forecast the long-term production of natural gas by region, and also globally. Both conventional and unconventional gas production are considered. Our results show that world natural gas production is likely to peak in the range 3.7 to 6.1 trillion cubic meters per year (tcm/y) between 2019 and 2060 depending on assumptions made on the size of the global ultimately recoverable resource (URR) of natural gas. A comparison of this paper’s forecasts with those from the scientific literature and from major energy institutes shows that the projection in this paper’s ‘high scenario’ can be seen as a likely upper-bound on future global natural gas production. To turn this upper-bound projection into reality, great efforts will be needed from the gas industry to discover more conventional and unconventional gas resources, and to make these recoverable. Full paper: https://www.sciencedirect.com/science/article/pii/S2352484720304200
... The second category includes world fossil fuel production estimates by geographic/political regions. For example, Al-Fattah and Startzman (2000) and Imam et al. (2004) forecast the gas production of Eastern Europe and FSU as a whole in their world natural gas production modelling. Evans (2011, 2009) have projected natural gas and coal production at the FSU region level. ...
Article
Full-text available
Detailed projections of the Former Soviet Union (FSU) fossil fuel production has been created. Russian production has been modelled at the region (oblast) level where possible. The projections were made using the Geologic Resource Supply-Demand Model (GeRS-DeMo). Low, Best Guess and High scenarios were created. FSU fossil fuels are projected to peak between 2027 and 2087 with the range due to spread of Ultimately Recoverable Resources (URR) values used. The Best Guess (BG) scenario anticipates FSU will peak in 2087 with production over 170 EJ per year. The FSU projections were combined with rest of the world projections (Mohr et al. 2015b), the emissions from the High scenario for the world are similar to the IPCC A1 AIM scenario. Supplementary information: The online version contains supplementary material available at 10.1007/s40789-021-00449-x.
... During the same period, Hubbert proposed a logarithmic growth curve called "Hubbert curve" (Hubbert, 1949(Hubbert, , 1956, which can describe the changes in oil production by analyzing the life cycle of fossil energy production in the United States. The curve model successfully forecasted the peak of U.S. oil production from 1967 to 1973 for nearly 10 years in advance, making "Hubbert curve" extremely famous, and many researchers have applied it to the forecasting of natural gas consumption (AI-Fattah and Startzman, 2000;Siemek et al., 2003;Wang et al., 2016). Berrisford (1965) believed that successful natural gas demand forecasting depended heavily on finding the best weighted model for current and previous temperatures. ...
Article
Natural gas consumption forecasting technology has been researched for 70 years. This paper reviews the history of natural gas consumption forecasting, and discusses the changes in forecasting horizons, influencing factors, and forecasting performance. According to the characteristics of forecasting models used in different periods, the history of natural gas consumption forecasting can be categorized into initial stage, conventional stage, AI stage, and all-round stage. The stage characteristics, typical models, advantages and disadvantages at different stages have been summarized. The review results show that, affected by the development of computer science and AI technology, short-term forecasting is the fastest-growing forecasting horizon, followed by long-term and medium-term. Additionally, long-term forecasting is mainly affected by production, population, and economic variables. Medium-term forecasting is mainly affected by economic and temperature variables. Influencing factors of short-term forecasting mainly depend on temperature variables, weather condition and date type. Furthermore, the statistical analysis of data characteristics, model characteristics and forecasting results presents that time series models are the best models for long-term forecasting. It has the lowest average mean absolute percentage error (1.90%) in long-term forecasting. To the medium-term and short-term forecasting, AI-based models present the best performance. Among them, artificial neural network models (2.21%) are preferred for medium-term forecasting, and support vector regression models (4.98%) are more suitable for short-term forecasting. Besides, this paper proposes a framework for model selection, and provides specific suggestions for future research directions.
... There are many indicators, shows that the future of the environment will be, unfavorably affected if the world dose's change, the energy production and consumption behaviors. [3,4]. ...
Article
Full-text available
Jordan like the other non-oil countries, spend significant amount of its budget on importing foreign energy. Also, the power demand increases due to, the population growth, in addition to hosting high number of refugees, from neighboring countries, and the enhancement of the citizen life style. The objective of this study, is to define the cooling load requirement, for the engineering building, at Mutah University, in order to analyze suitable cooling system, to cater the required cooling load. Energy analysis carried out, using the climate variables tabulated data, and Cooling Load Temperature Difference (CLTD), and Cooling Load Factor (CLF). The required power, to drive the cooling system was defined. Climate data shows that the average temperature was 33.2 °C, 32 °C and 33.3 °C in 2015, 2016 and 2017, respectively. The maximum temperatures was 38 °C registered in 2015. In this study the implementation of renewable energy, was considered by analyzing the viability of using, photovoltaic system, to provide power for the cooling system. Two options, to provide power for the system analyzed, the first option is supply the power from the grid, and the second option is, to supply the required power using photovoltaic power plant. The results show that, the cooling load of the building is 560 kW, and the required electrical power, to operate the cooling system is 224 kW electrical power. In addition, the results show that, the cost of power from PV system is, 40 % less than the cost of power from grid for the first 10 years. In addition, the payback period for the designed system is 5 years.
... Today, this bell-shaped curve and its mathematical equation are famously known as the Hubbert curve and Hubbert model . Due to society's concerns about a potential shortage of non-renewable fossil fuels, numerous studies began to forecast the long-term production of fossil fuels using bell-shaped curve-fitting models (e.g., Al-Fattah and Startzman, 2000;Imam et al., 2004;H€ o€ ok et al., 2011;Wang et al., 2011Wang et al., , 2013a2013b, 2018Maggio and Cacciola, 2012;Saraiva et al., 2014;Harris et al., 2018). ...
Article
Rare earths are increasingly important in modern industry. It is thus critical for both users and policy makers to understand the future production of rare earth resources. This paper presents a quantitative analysis of the long-term production of world rare earths using a flexible bell-shaped curve-fitting model. The results show that world rare earth production may increase for many decades. However, the production rate, which is key to understanding the balance between the supply and demand of rare earths, may differ sharply with variation in the value of ultimately recoverable reserves and the model's curve shape. Our highest projected for peak production and peak year are 1.28 Mt (rare earth oxide equivalent) and 2115 respectively, and our lowest projections suggest that production will peak at 0.4 Mt in 2118. Given the many potential constraints on production, this paper suggests that our lowest projection may best represent the actual production pathway. Finally, the paper suggests measures to deal with a potential imbalance between supply and demand from both the supply and the demand side.
Article
Full-text available
The amount of ultimately recoverable resources and/or timing of peak production have been the central purpose of numerous studies. One broadly applied method is Hubbert modeling, subsequently extended as multicyclic Hubbert modeling. This paper explores a modification to conventional multicyclic Hubbert modeling that we term “cycle‐jumping” wherein the overlap of multiple curves is limited and explicitly accounted for. The model is designed in a way that each curve is described by the same three parameters as a multicyclic Hubbert model, and every two curves are connected through an explicit transition. The transition width indicates the time of the shift from one curve to the next and is controlled by a weighting parameter for the respective curves. Cycle‐jumping provides a superior data fit compared to the conventional cycle‐addition model and, more important, reflects historical production data more realistically as socioeconomic and political factors important to resource production vary in time. Recommendations for Resource Managers • Conventional multicyclic Hubbert modeling poorly reflects transitions in production trends. • Cycle‐jumping with a finite transition period practically and mathematically provides a superior model for historical resource production by limiting the overlap of multiple curves. • Cycle‐jumping with a finite transition period reflects more realistically the production profile affected by external factors including capturing inherent asymmetry in different cycles.
Article
Full-text available
The modeling of Natural Gas (NG) demand differs significantly from the demand for electricity in terms of the determinants that affect it, as all fields of economic activities in a modern economy are directly related to electricity but not to NG. But NG is the second energy type after electricity used in all countries in percentages greater than 10% in average terms. NG is going to be installed in the Region of East Macedonia-Thrace (REMTH) the next years. So, we consider it is worth to predict the NG demand in REMTH using eight deterministic forecasting models. In order to do it we used a dataset of 20 years concerning two Greek regions to which the NG is used that period and through them we built the eight forecasting models aiming to find the NG demand in the REMTH. In order to evaluate the reliability and accuracy of them we used four types of statistical errors, Mean Error (ME), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and Mean Absolute Scale Error (MASE). These are the most widely used measures of evaluating the accuracy of deterministic predictive models, due to their advantages of scale-independency and interpretability. When each of them is used alone has the significant disadvantage to produce infinite or undefined values for zero or close-to-zero actual values. In order to address this disadvantage, we propose a way to use the same time all of them measuring the accuracy of a model used to forecast the demand of Natural Gas in the Greek region EMTH. The innovation of this paper is that for NG demand forecasting were used seven different models and they are evaluated regarding their reliability /accuracy using five types of residuals or statistical errors
Article
In this study, we undertook the risky task of predicting the world's future supply of and demand for petroleum liquids [crude oil and natural gas liquids (NGL's)] to the Year 2050. On the basis of earlier work, we choose to forecast future supply using the Hubbert model, which has advantages of simplicity and availability of data. Although this approach has several known inadequacies, it has withstood the test of time in numerous cases, and we consider it a fairly reliable supply predictor. We also used the Hubbert model to forecast demand. We found no precedence for this use and currently regard its accuracy as speculative. Nevertheless, some signs indicate that it too may develop use over time as a reasonable predictor. Our study showed that world supply (and demand) for petroleum liquids is currently at a peak and will soon decline. We examined supply and demand by region. Some regions, such as the Middle East, will produce a surplus of crude throughout the forecast period. Other regions, most notably the Far East and North America, will remain importers. Still other regions, such as Africa, that are now exporting will soon become importers.
Article
Because gas and oil are exhaustible resources, the discovery and production history of these fuels in any particular area must be characterized by a beginning, a period of increase, a period of decline, and ultimately, an end. In this sequence, the most significant dates are neither those of the beginning nor of the end, but that of the transition between the period of increase and the period of decline. The petroleum industry in the conterminous United States is by now sufficiently advanced in its evolution that several lines of evidence are consistent in their indication of the approximate degree of that advancement. One suite of diagnostic data comprises the cumulative production, cumulative proved discoveries, and proved reserves. For more than 40 years the curve of cumulative production of crude oil has closely followed that of cumulative discoveries, with a time lag of about 10-12 years. The peak in the rate of proved discovery occurred about 1956. The peak in proved reserves, which should occur about midway between the peaks in the rates of discovery and of production, actually occurred at about 1961. The peak in production rate should occur in the late 1960's. Another suite of data is provided by successive studies made by the Petroleum Administration for War and by the National Petroleum Council, in which the oil discovered has been allocated to the years of discovery of the producing fields. These, when corrected to an estimated ultimate growth, indicate that by the end of 1966 about 136 billion bbl of producible crude oil had been discovered. The rate of discovery per year, averaged for successive 5-year periods, reached a peak of 3.57 × 109 bbl/yr during the period 1935-1940, and has declined subsequently to a present rate of less than 2 × 109 bbl/yr. Another method of analysis is that proposed by A. D. Zapp (1961, 1962), of relating discoveries per foot of exploratory drilling to cumulative footage. By the end of 1966 it is here estimated that about 1.52 × 109 ft of exploratory drilling had been done. The discoveries per foot, averaged for each successive 108 ft, reached a peak of 276 bbl/ft during the third 108-ft interval, which included the discovery of the East Texas field. Subsequently, the rate of discovery per foot has declined exponentially to a present average rate of about 35 bbl/ft. These several lines of evidence are consistent indicating that: 1. The peak in time-rate of crude-oil production will probably occur near the end of the 1960 decade, and the ultimate amount of oil production will probably be less than 200 billion bbl. 2. The peak in the rate of natural-gas production will probably occur in the late 1970's, with ultimate production between 900 and 1,200 trillion cu ft. 3. The ultimate cumulative crude-oil production from fields already discovered in the conterminous United States probably represents between 68 and 85 per cent of the total crude oil ultimately to be discovered. 4. The evidence available offers scant hope that recently postulated future discovery rates in excess of 80 bbl/ft, with attendant discoveries of additional hundreds of billions of barrels of crude oil, during the next 2 or more billion ft of exploratory drilling, will ever, in fact, be realized.
Article
Growth in estimates of recovery in discovered fields is an important source of annual additions to United States proven reserves. This paper examines historical field growth and presents estimates of future additions to proved reserves from fields discovered before 1992. Field-level data permitted the sample to be partitioned on the basis of recent field growth patterns into outlier and common field set, and analyzed separately. The outlier field set accounted for less than 15% of resources, yet grew proportionately six times as much as the common fields. Because the outlier field set contained large old heavy-oil fields and old low-permeability gas fields, its future growth is expected to be particularly sensitive to prices. A lower bound of a range of estimates of future growth was calculated by applying monotone growth functions computed from the common field set to all fields. Higher growth estimates were obtained by extrapolating growth of the common field set and assuming the outlier fields would maintain the same share of total growth that occurred from 1978 through 1991. By 2020, the two estimates for additions to reserves from pre-1992 fields are 23 and 32 billion bbl of oil in oil fields and 142 and 195 tcf of gas in gas fields. 20 refs., 8 figs., 3 tabs.
Article
A quantitative analytical method, using a spreadsheet, has been developed that allows the determination of values of the three parameters that characterize the Hubbert-style Gaussian error curve that best fits the conventional oil production data both for the U.S. and the world. The three parameters are the total area under the Gaussian, which represents the estimated ultimate (oil) recovery (EUR), the date of the maximum of the curve, and the half-width of the curve. The best fit is determined by adjusting the values of the three parameters to minimize the root mean square deviation (RMSD) between the data and the Gaussian. The sensitivity of the fit to changes in values of the parameters is indicated by an exploration of the rate at which the RMSD increases as values of the three parameters are varied from the values that give the best fit. The results of the analysis are as follows: (1) the size of the U.S. EUR of oil is suggested to be 0.222 1012 barrels (0.222 trillion bbl) of which approximately three-fourths appears to have been produced through 1995; (2) if the world EUR is 2.0 1012 bbl (2.0 trillion bbl), a little less than half of this oil has been produced through 1995, and the maximum of world oil production is indicated to be in 2004; (3) each increase of one billion barrels in the size of the world EUR beyond the value of 2.0 1012 bbl can be expected to result in a delay of approximately 5.5 days in the date of maximum production; (4) alternate production scenarios are presented for world EURs of 3.0 and 4.0 1012 bbl.
Energy Resources Publication 1000-D, Natl. Academy of Science/Natl
  • H U B B E Rt
H u b b e rt, M.K.: " Energy Resources, " Publication 1000-D, Natl. Academy of Science/Natl. Research Council (1962).
I n t e rnational Energy Outlook DOE/EIA-0484, Off i c e of Integrated Analysis and Forecasting
" I n t e rnational Energy Outlook 1998, " DOE/EIA-0484, Off i c e of Integrated Analysis and Forecasting, U.S. Dept. of Energ y, EIA, Washington, DC (April 1998).