THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS. SOCITY. CHINA 105
A Petroleum Life Cycle Model for the United States
with Endogenous Technology, Exploration, Recovery, and Demand.
BY
P&ll. Davidsen, mag. art., Associate Professor,
Institute for Information Science , University of Bergen, 5014 Bergen - Universitetet, Norway.
(1986: The System Dynamics Group, Sloan School of Management,
Messachusetts Institute of Technology, Ma. 02139, USA).
John D. Sterman, Ph. D., Associate Professor,
The System Dynamics Group, Sloan School of Management,
Massachusetts Institute of Technology, Ma. 02139, USA.
George Richardson, Ph. D., Associate Professor,
Wheaton College, Norton, Ma. 02766, USA.
ABSTRACT
‘This paper describes model of the life cycle of the petroleum resowce in United States. Expand-
ing on prior system dynamic models of petrolewn resources, the model endogenously generates the
complete life cycle of the resource. The model treats endogenously the petroleum demand, the
development of technology for and investment in exploration and recovery, the discovery and
production of petrolewn and the development of petroleum substitutes. With only two exogenous
veriables (GNP and the international petroleum price), the model is able to portray the evolution of
‘the petroleum resowe, and the associated industry, starting in 1870. The correspondence between
simulated and actual data is exemined through a variety of statisticel messures. The model is used t9
show how the interplay of technological progress, depletion, and the development of substitutes
create the lifecycle by altering the dominance of the feedback processes in the system. A full
documentation may be found in (Davidsen, 1987).
‘The model constitutes the besis for more comprehensive studies in the following areas of concern,
all elated to the management of depletable resources:
1. Adaptation of the model to different kinds of depletable resources, and generalizations to obtain
amodel portraying the generic structure of the life cycle of adepletable resource.
2. Analysis and evaluation of current practices in the management of depletable resources.
3. Design and evaluation of altemative strategies for the management of such resources.
‘The model seems furthermore to be a promising tool for teaching resource management.
PURPOSE AND OVERVIEW
This paper describes a model of the life cycle of petrolewn resources in the United States. Expand-
ing on prior system dynamic models of fossil fuel resources (Naill, 1973; Naill, 1977; Backus et.
al., 1979; Choucri, 1981; Sterman, 1981; Stennan and Richardson, 1985) the model endogenously
generates the complete life cycle of this resource. The model synthesizes the perspective of several
specific diciplines, such as geology, technology and economics. It integrates exploration, produc-
tion, pricing , demand, imports, and the development of substitutes. Finally, the model emphasizes
the impact of delays in both the physical processes and the information ! decision making processes
of the system.
‘With only two major exogenous variables (GNP and the intemational petroleum price), the model
is able 0 portray the evolution of the petroleum resoue, and the associated industry, starting in
1870. The conespondence between simulated and actual data is examined through a variety of
106 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA
statistical measures. The model is used to show how the interplay of technological progress, deple-
tion, imports and the development of substitutes, creates the lifecycle by altering the dominance of
‘the feedbeck processes in the system. The life cycle is characterized by exponential expansion of
the petroleum industry, driven by economic growth, developing geological knowledge, and techno-
logical progress, followed by a transition to decline, driven by depletion, rising real costs of explo-
tation and production, and ultimately to the development of substitutes.
The model is intended to provide a realistic "micro world" in which the geological and technical
parameters are known end can be varied to portray alternative scenarios. The model can then be
‘used for a variety of puposes:
1. The model will be used to generate synthetic data for the modeling and evaluation of resource
en methods (Stennan and Richardson, 1985; and Stenman, Richardson, and Davidsen
1987).
2. Dees and mutually consistent forecasts of production, exploration activity, and costs can be
developed.
3. Policies regarding e.g. price controls, taxes and import fees can be evaluated in a rich dynamic
environment which represents the feedbacks important in the real system.
4, The model is reasonably tansparent and offers opportunities to teach resource management,
dynamic modelling , and principles of feedback.
‘The structure of the model is described , followed by a discussion of the parametric assumptions,
base case behavior, and potential applications.
‘The petrolewn industry begen in eamest in 1859 with Colonel Drake's famous well in Titusville,
Pa.. A model, such as the one described here, which portrays the full life cycle over 130 years of
history and beyond , must meet certain requirements a short tenn forecasting model does not:
First, it must be @ structural model. It should represent the physicel end causal structure of the
processes modeled , as opposed.to a model based on historical correlations. Non-linearities and
constraints may alter the historical correlations in the future. Physical delays, such as the time
required to develop an oilfield or build a synfuel plant, should be represented explicitly.
Secondly, it should be:a behaviourel model, portraying the information available to actors and the
procedures they use to process itand arrive at decisions. The petroleum system is characterized by
imperfect information, uncertainty, and distributed decision making. If the model is to respond to
changes in the environment in the same way that real ectors do, this bounded ‘rationality should be
incorporated (Simon, 1947, - 57, - 79, - 82; Hogarth, 1980; Morecroft, 1983; - 85).
‘Thindly, the model should generate its behavior endogenously. The exploration and production pro-
cess is tightly interconnected with energy price, demand, import, substitution, and technology. A
change in one part of the system may have ramifications throughout. A model that relies on exogen-
ous variables is likely to produce inconsistent results as the feedback effects are ignored. A model
‘that generates the petroleum life cycle endogenously constitutes an internally consistent theory that
is subject to analysis, refutation and revision (Bell and Senge, 1978).
In addition to these general considerations, a model of petroleum resources to be used in forecest
evaluation should include the following specific features as endogenous components:
1, Demand, import, and substitution. Petroleum demand is sensitive to price. As the prices rise, the
demand for petroleum will be depressed, and the production of substitutes ("backstops" [Nord-
havs, 1973]), such es synfuels, will be stimulated. If the price on domestically produced petol-
eum rises above the import price, import is indicated. The pattern of demand, import and substi-
tution will have a strong influence on production and investment in domestic exploration. Delays
in response of demand and in the development of the backstop industry should be made explicit.
2. Depletion thmpngh exploration and production... The total quantity of petroleum initially in-place
is finite. As itis discovered, produced, and consumed, the quantity remaining inevitably de-
clines, and the marginal cost increases, ceteris paribus. Though improving technology may off-
set depletion and cause the real price of petroleum to decline, ‘the limited nature of the resource
dase and its depletion should be treated explicitly.
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 107
3. Technology. The ultimately recoverable resource depends heavily on the recovery fector. Only
30-40 per cent of the oil in place can be recovered economically with cuzent technology, but
the fraction recoverable has been rising, and may rise substantially in the future. Similarly there
is a development of exploration technology. The effects of investments in technological devel-
opment should be treated explicitly.
4. Economic incentives; petroleum prices. Economic incentives (primarily determined by the
petroleum prices) Playa. large role in determining proved reserves, exploration, and produc-
tion. Petroleum that is subeconomic at $10 per barrel may be highly profitable at $30 per
barael. Regions that were not even considered for exploration may be prime candidates for test
wells at a higher price. Because the price has a strong influence on the incentives for explore-
tion and development, it should be modeled explicitly. The effects of production costs, supply
and demand, market imperfections and imports should be incorporated.
‘The sectors of the model are exhibited in figure 1.
THE MODEL
EXPLORATION AND PRODUCTION
The model divides the total quantity of oil-in-place into three basic categories; as yet undiscovered
petroleum, identified reserves and cumulative production (figure 2). Within these broad categories,
several finer divisions are portrayed (figure 3). The disaggregation of the resource base follows
standard resource classification shown in the McKelvey box (USGS, 1976) (figure 4). Successful
exploration shifts the boundary between identified and undiscovered resources to the right; im-
provements in technology or increases in the real price of oil shifts the boundary between economic
and subeconomic resources towards the bottom. Production shrinks the reserve base.
In this section, the physical structure of the exploration (discovery) end production (recovery) of
the resource, is described. Of major concer are the determinants of the productivity of investnents
in exploration and production.
‘The productivity of investment in exploration is negatively influenced by the discovery rete (figure
‘S): Suppose the discovery rate is increased. Then less remains ‘o be discovered with current tech-
nology, and the productivity of further investment in exploration is reduced. Itis assumed that the
‘yield from exploration is exponentially decreasing with the cumulative footage drilled , and that the
footage drilled per $ invested is constant (Hubbert, 1969, - 75; Hall and Cleveland, 1981). The
reduction in productivity feeds bat to the discovery rate implying a reduction in the discovery
potential provided by any given level of exploration activity.
‘The productivity of investment in production is influenced in a similar manner by the rete of pro-
duction (figure 6): Suppose the production rate is increased. Then less remains to be recovered.
‘Thus the productivity of investment in production is reduced, feeding back to the production rate 10
reduce the production potential, provided by the investment in production. The production potential
constitutes an upper limit for the production rate, a rate which may thus be reduced. Note that the
technically recoverable resource remaining constitutes an upper limit for the rate of production as
well,
‘These underlying physical structures tend to stabilize the discovery and recovery of petroleum. As
Jong as there is a demand for petroleum, the productivity of the investments will be exponentially
reduced a3 more of the resource is discovered and more of the identified reserve is recovered. The:
consequence is a tendency to slow down the rate of depletion, ie the rate at which the system.
approaches its equilibrium. In addition to demand, there are only two more factors that may influ-
ence the development towands equilibrium in each sector, changes in exploration effort and chang-
es in technology. Increesed investment in exploration increases the discovery rate. Better techno-
logy improves the productivity of investments by making more of the petroleum available, - i.e.
discoverable and recoverable.
108 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA
‘These two sectors are interrelated physically since exploration provides the identified reserve,
which constitutes the basis for production. Progress in exploration also has an impact on the pro-
ductivity of investment in production: Suppose that less remains undiscovered due to more exten-
sive exploration. In that case, the production is allowed to take place in more demanding geo-
structures. Now, if production takes place where exploration has recently taken place, the produc-
‘tivity of the production will then comespond to the current productivity of the investment in explo-
ration. The recovery of a field however, normally takes place somewhat after the field has been
discovered. As exploration progresses, the technically recoverable reserve discovered accumulates.
‘Thus the productivity of investments in production is lagging the productivity of investment in
exploration, coresponding 10 this reserve.
INVESTMENTS
Investments are made to build up an exploration potential (capasity) and « production potential (fig-
ure 7 and 8). These investments are determined by the demand for petroleum and the petroleum
price. In addition, the productivities of such investments, a3 they compare to the market price, play
an important role in the investment decisions.
Investment augments the capital stock for exploration (drill rigs). The time required to allocate
funds for, acquire, and conduct the exploration activity is represented explicitly. An average lag of
4 years is assumed. Once successful exploratory wells have been drilled, there is a further 1 year
averege lag in the development of production wells.
Investments in both exploration and production are adjusted to the perceived productivities of such
investnents, - though have to be justifiedby the market price. The breakeven prices required to jus-
tify exploration and production are therefore compared to the prevailing market price. If the market
“price falls below the required prices, investments ‘ill be curtailed. The perceived productivity of
investments in exploration is assumed to leg the real productivity by 15 years on the averege.
TECHNOLOGY
Petvolewn technology may be divided into exploration and recovery technology (figures 7 and 8).
‘Technological improvements increase the availability of petrolewn through exploration and pro-
duction, and thus improves the productivity of the investments. The current level of sophistication
in exploration technology is expressed by the fraction of the total petroleum resource that can tech-
nically be discovered. The level of sophistication in production technology is expressed by the frac-
tion of the identified reserve that can technically be recovered. Both exploration ard recovery tech-
nology are endogenously improved by investments. As these fractions approach their maxima, the
marginal effect of further research and development diminishes (technological saturation).
What may technically be discovered and recovered at any point of time, and thus the ratio between
gross and net yield from exploration, is determined by the technological development. The net yield
influences the demand for exploration and the calculation of the unit exploration expenditures, upon
which the petroleum price is based.
‘The origin of technological progress lies in investments from revenues, i.e. the product of petrol-
eum price and production: A fixed fraction of the revenues is assumed allocated to research and
development. Because of the technological saturation, the marginal prod.uctivities of these invest-
ments are declining. There is a6 years average third order delay in technological research and
development. The split of this investment between the two technologies is subject to changes over
time (figure 11). Of primary concem is the exploration technology, as the exploration creates the
basis for production. Gradvally, a3 this technology approaches its maximum level of sophistica-
tion, the emphasis is shifted from exploration to production technology so as to utilize the identified
reserve.
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 109
Depending upon the revenues, technologicel improvements allow for exploitation according to the
petroleum demand, while compensating for the decline in the productivities of investments in
exploration and production. Therefore, the stwonger the technological progress is, the more aggres-
sively the depletion effect is compensated , - and further investments made in technology. When, on
‘the other hand, the production is falling, less is contributed to technological development. There-
fore the fall in productivity is compensated less aggressively, promoting a further decline in the
production ret.
As long as the production is growing, and the price is kept relatively stable by investments in tech-
nology, the production rate predominantly determines the revenues and the investments in techno-
logical resesrch and development. If the production levels off, this may be compensated by an
increase in the petroleum prices, in which case the technological progress is dominated by the
changes in price. Should the production decline, smplified by technological stagnation, then there is
acall for a substantial increase in price t sustain the revenues and the investments in technology.
Note that if the production sctually levels off and starts declining, then the impact of a change in
technological compensation is determined by the timing of the peak in production: A large impact
thay result from production peaking at an early stage in the technological development, because
then the contribution from technological development is relatively dominating. When the technolo-
gical saturation sets in, the effect of such a change will be less. Note also that the progress in
exploration technology generally diverges from the progress in production technology and that the
exploration costs and the production costs axe not the same (exploration costs generally being
substantially higher than production costs), so that the contibution from each of them may differ
significantly.
PRICE AND DEMAND
‘The demand for petrolewn is caused by our economy being petroleum intensive and is determined
by the exponential growth of GNP ( figure 10). Domestic exploration and production is in demand
to the extent that substitutes, provided by import of natural petroleum or production of synthetic
petroleum, are not available at lower that breakeven costs.
‘The demand for domestic production is complemented by the tendency to import (the indicated im-
port). A rising tendency, reduces the production pressure. The tendency to import is determined by
the ratio between the intemational (import) price and the exploration and production expenditures.
Note that the unit exploration expenditure is the average exploration cost per barrel essociated with
the recoverable reserve. Thus the tendency to import is smoothly effected by changes in the produc-
tivity of investments in exploration. The actual import endogenously covers the residual demand,
Le. the demand not satified by domestic production.
‘There sre two different ways in which the demand for petroleum may be influenced by the petrol-
eum price over time: First of all the demand is reduced by rising prices causing the energy intensity
of GNP to decline. Furthemnore, a synthetic petroleum industry may be justified by higher prices.
‘There are substantial delays associated with the impsct of price on demand: It tkes 15 years on the
average to adjust the energy intensity, considering the potential for retrofitting existing capital, as
the life of energy consuming capital is 20 years on the average (Coen, 1975; Stennan 1981). To
build a synthetic petroleun industry takes 9 years on the average. Synthetics, which represents a
perfect petrolewn substitute, is assumed to cost $50 per barrel.
As long as the domestic petroleum price is not dominated by the international price or the price of
synthetics, three factors determine the price (figure 12);
~ the costs associated with the exploration and production of petroleum;
= the demand for pevoleum; and
- the supply of petroleum.
Before 1953, the US domestic price is considered dominating the international market, go that the
price is endogenously determined by the cost of exploration avd production. From then on, te
110 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA
domestic production is protected by import quotas until 1972. Yet the price is gradually more influ-
enced by the fall in the international price. In 1971 a price control is introduced. This control is
effective until 1981 contributing to avoid a windfall profit from dramatically rising international
prices, During this period the domestic price approaches the international price from its base level
which is predominantly detemnined by costs. After 1981, no controls ave in effect, and the domestic
price is completely determined by the intemational price.
The impact of the demand on price is relatively straightforwand: Suppose the demand for domestic
production increases. Then so does the petroleum price because the petroleum market tends to
develop towards a sellers’ market. The impact of the petroleum supply is the opposite one: Suppose
the supply is increesed. The price is then mduced because the petroleum market tends to develop
towards a buyers’ market. The supply of petrolewn is represented by the production potential. Note
that the effectof supply and demand on price is relatively limited: The oligopolistic petolewn.
market is characterized by short tem adjustments of the potential production (supply) according to
the demand and the merket price. It is therefore not very common that the price is affected by an
abundant supply. Shortages are primarily result of depletion, insufficient investment in explore-
tion, and insufficient supply of imports or substitutes. Urier a dramatic upward pressure on price,
‘there is a tendency to introduce price contols to protect the petolewn conswning industry and
avoid windfall profits..As the price approaches the international level, such regulations axe
abolished.
The demand for petrolewn exploration originates from three different requirements (figure 9), ie.
b,
- keep up with current production;
- maintain a reserve recoverable, adequate t keep up with production; and
- adjust the current recoverable reserve in accordance with the expected growth in demand.
‘The amount to be produced is simultaneously required substituted by recoverable petroleum. The
reserve smaust be corrected to an adequate level, conesponding to the current rate of production. Any
discrepancies from this level is phased out over a relatively long period of time (15 years on the.
average). Furthermore, an adjustnent must be made according to the expected growth in demand.
This adjustment is immediate, but is based on the forecast of a recognized trend in growth. There is
assumed to be aS years averege delay representing the time to observe, perceive, and recognize this
tend, and en additional S years to average the wend as a basis for forecesting. As the growth trend
iz calculated fram past demand, theze adinztmen‘s of the reserve aie hoth based upon the demand
for production, and the current reserve technically recoverable.
‘We may conclude that the demand for domestic production, when transformed into a demand for
exploration, is amplified for two different reasons: First of all the recoverable reserve must be
established and maintained. Secondly the gross yield from exploration must be considerably larger
than the required net yield, due to the inadequacy of the production echnology.
‘The price on and demand for petrolewn is determined as swnmerized in figure 13 (where the empli-
fying effect of the inadequate recovery technology on exploration, the effects of supply and de-
mand on price, and the short term adjustments of investments eccording to productivity and price,
aye all left ous). The bold part of this figure represents the system considered without a technology
sector. The remaining represents the impact of technology. Two majox feedback processes tend to
stabilize the system, respectively via;
~ the influence of exploration and production costs on the petrolewn price; and
- the impect of import tendencies on the demand for domestic production.
As the wit costs increase due to depletion, these processes cause the demand for petroleum to stag-
nate. Thus the rates of discovery and production are moderated. Due fo the delayed, exponential
nature of this effect, its impact is initially insignificant. Therefore exploration and production nay
incease with the exponential demand created by the growth in GNP. This gives rise to technologi-
cal investments and progress. Provided this progress more that offsets the inpact of depletion on.
productivity, the petroleum price is falling. This amplifies the rise in demand , production and.
technological progress through positive feedback. When the technological saturation sets in, the
progress can no longer offset the depletion. The rise in costs, price and imports, forces the demand
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 111
for domestic prod uction into decline. This reverse tendency is again amplified by the positive tech-
nological feedback. (Note that there is a negative feedback, via the impact of the petroleum price on
revenues, thet may dominate the technological progress during a transition period as the production
levels off).
THE DYNAMIC MODES OF THE SYSTEM
INTRODUCTION
In this section the dynamic behavior of the model vill be explained and related to historic records.
The system exhibits two quite different dynamic modes of behavior, both driven by the exponential
growth in the GNP:
One of these modes originates from the physical consequences of petrolewn exploitation, - i.e.;
~ the discovery of petroleum, leaving less of the total resource to be identified a3 & reserve in the
future; ond
- the recovery of petroleum, jeaving less of the identified reserve for future recovery.
‘The result is an exponential decline in the productivities of exploration and production. If the in-
vestments were kept constant, and there was no technological innovation, this would cause the dis-
covery rate and the production rate to approach zero exponentially. In this case the mode of beha-
‘vior is an effective contraction of exploration and production. Under such circumstances, and pro-
vided there are no price constraints, the petroleum price would rise acconding to the declining pro-
ductivity. Thus the demand would no longer follow the exponential growth in GNP, but gradually
level off.
To the extent that the price mechanism does not offset the growth in GNP, there om two different
‘ways to compensate for the contraction of exploration and production, in order to accomodate the
residual growth in demand created by GNP. That is to increase investments in;
- exploration and production; and in
~ research and development of the two technologies.
Note that these two kiruls of compensating measures act through very different processes. Direct
investments tend to increase the volume of the exploration and production activity, Le. the rate at
which the exploration and prodvction frontier is extended into the geo-structure. Lerger investments
‘therefore tend to reinforce the contraction. Investments in research end development tend to make
exploration and production more efficient. Technological progress may allow for the frontiers to be
extended ata lower rate, still with the same net vieli Thus to compensate for deteriorating prodnc-
tivities, one may invest in technology to complement, and even substitute 1u1 aevesuueues pivmue
ing depletion.
Even steonger echnological investments may more than cancel out the effect of depletion and acta-
ally cause the productivities to increase. In that case, we are facing an expanding mode of behavior
with the following charactetistics: The petrolewn price will be falling as long as the technological
development is "on top of" depletion. Note that this in time will reinforce the petrolewn demand, - @
development that will emplify the pressure for technological progress. To the extent that there ‘is an.
“adequate response to this pressure, the exploration and production may continually satisfy the”
demand without increasing unit costs. If the technological response tums out to be inadequate, then
cwe experience the contraction.
‘The dynamics of the life cycle of the US petrolewn resource is characterized by expansion followed
by contraction. There are however a set of factors that tend to modify this behavior. Four of the
most prominent ones are;
- the ilentified reserve, acting a3 a buffer between exploration and production;
~ the delay in recognizing the marginal exploration cost, acting as a buffer between cost and invest:
ment,
112 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA
- the delay in averaging the exploration expenditures, acting a3 a buffer between cost and price,
and
- the regulations, scting as buffers between the US's and the international petroleum economy.
‘These buffers may mask the transition from expansion. to contraction, and tend eventually to ampli-
fy the contraction. Dué to its significance, the transition will be designated az a particular mode of
behavior.
‘THE EXPANSION (1870 -1945)
The dynamic mode of the petroleum life cycle is a typical expansion, end hes the following pmpex-
ties (figures 14 to 23): Historically the petroleum price veried substantislly in the beginning (figure
23). But by-and-lorge it declined with a decreasing margin until and was fairly stable thereafter
{until after World War I). The demand for petrolewn wes detennined by the fact that it was pie-
dominantly used es a source of light, - only later to be recognized 3 an energy source suitable for
heating, production, and transportation. Gradually, petrolewn was established as a mature sowrce
of energy with a relatively stable market share (1950 - }. With the increasing market penetration that
petrolewn had during the first part of this century, the growth in demand (figure 21}, exploration
atl production exceeded the growth in GNP. The production exhibited a growth pattern core-
sponding to the demand (figure 22), - and elloved in eddition for export. Note that duzing most of
‘this tine, USA was a net exporter of petroleum.
‘The energy intensity in the model is calibrated to represent the prevailing intensity during the last 35
years. Therefore an exogenous factor represents the smooth transition from other sources of energy
‘to petroleum. This causes the petroleum demand to grow according to the historic demand (figure
21),- Le. faster than the reference growth for most of this period. In addition to that, this growth is
endogenously amplified by a declining petioleum price.
In the model, the production grows exponentially (figuze 14, 16) end satisfies the domestic demand
and the petroleum export. The growth in net yield from exploration is also exponential and satis-
fying during this period (figure 18). The technically recoverable reserve is developing normally
(figure 17), acconding to the growing rate of production, until 1928. Thereefter, an excess reserve
accumulates over time bringing the potentisl production from reserves well above the actual produc-
tion (figure 16), in spite of the fact that the investments in exploration axe adjusted to changes in
costs. The accumulation originates from. the substantisl delay in recognizing the decline in unit
exploration expenditures, causing overinvestments to be made. Both kinds of technology are devel
oped at an exponential rate so 23 % more than offset the impect of depletion on productivities figure
17). The progress in discovery technology reaches its maximum effect (the inflection point of the
frection discoverable) in 1931. The effect of the technological progress in production is consider-
ably more moderate. The technologicel progress accounts for an increasing productivity of invest-
ments, causing the petroleum price to decline from its initial value until 1930, when it levels off
(figure 20) (The price exhibits a fluctuation which is a litle more moderate than what was experi-
enced historically, and is systematically slightly higher after 1930 (figure 23)).
Inonder to sustain the technological development that makes it possible to satisfy demand, ever
larger investments must be made. Because the technologicel investments are financed 43 @ fraction
of the revenues and the price is declining sssymptwiically, it is the exponential growth in production
‘that facilitates further growth Uuough the positive feedback process described under “Technology”.
‘THE TRANSITION (1945 - 1981)
‘There axe upper limits os to how much of the petroleum that may be discovered and mcovered. As
‘these limits are approsched,, the technological prog :eas is characterized by saturation, and the mar-
ginal effects of the investments in technology ar veduced. Thus the need for investments in tech-
nology, just to compensate for this decline, is increased exponentially. Recall in addition, that the
“partial effect of a constant exploitation is to reduce the productivities exponentially, and that the
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 113
growth in petrolewn demand is also exponential. The growth in technology investments should
therefore compensate for a hyper (tripple) exponential decline in the return on these investments.
Sooner or later, we may conclude, the technological innovation are not adequate to withstand the
effects of exploration and production. Thus we approach a contiection, primarily characterized by
declining productivities. But before the contrection is fully established a3 a mode of behavior (from
1981 on), a lot happens, - both historically and in the model:
Itis not surprising that the historic records on price, demand and production hardly show any signs
of the oncoming contraction during the transition mode. As already pointed out, this wansition is
being masked by intemal and external buffers (delays, reserves and regulations). The import(195.
~ 73) and price (1971 - 81) regulations are the only ones that may be:tréced empirically. Note that
the effects of the oil-embargo (1973) and the Iran - Irak war (1978): withizespect to importand
price, cannot be claimed to reflect domestic exploration end production realities. It is important to
notice however, that these events coincide with the final stages of the domestic tansition from.
expansion towers contraction. What historically happened was the following:
, ve
After World War II, the domestic petroleum price rose by more than $ 4 per barrel (figure 23). The
Price was then fairly stable, but drifted during the import regulation towards the intemational price,
which fell below the domestic price in 1950, - and reached a local minimum around $ 6 in 1972.
‘The price control then prevented the domestic price from rising 3 dramatically as the intemational
price during the two crises in the ‘70s. As the produced and imported petroleum reserves were
being depleted, the pressure from the international market forced the domestic price up towards the
intemational price (which was practically attained in 1981).
‘The growth in petroleum demand was exponential until 1973, stagnating only slightly (figure 21).
It was then inflected significantly, leveled off, and started falling.
‘The domestic production initially followed the same trend as the demand , but was supplemented by
a substantial growth in imports during the import regulation period (1953 - 73) a3 USA tumed from.
being a net exporter to become a net importer of petrolewn (figure 22). The domestic production
wes also supplemented by the Alaskan production which gradually grew and amounted to more
than 11% of the total domestic supply in 1986. Consequently, the production in the lower 46 states
contracted more significantly than the petroleum demand during this period, and fell substantially
during the period 1973 - 81.
Ttis convenient to start off the description of the transition mode of behavior, a3 portrayed in the
model, by pointing out the early vaming indicated by the inflection of the fraction discoverable in
1931 (figure 17). This represents the first sign that the positive, compensating feedback loop, pro-
moting the technological development, looses its dominance. Note however that the progress in
production technology has an accelerating effect up until 1967, when also the fraction recoverable
inflects due to saturation. But because the unit exploration expenditure at this point of thne is con-
siderably larger than the production costs, the contraction is triggered. One could then expect a. con-
‘traction scenario; - with lower productivities, higher costs and price, reductions in demand and
investments and a decline in exploration and production. But it takes a very long time to establish
this mode of behavior:
The petroleum price is relatively stable until 1973 (figure 20). Itdoes not reflect the post-war
increase, a3 the model does not represent the price regulations prevailing during the war. The
subsequent import and price regulations are however represented exogenously in the model. They
ave considered to be imperfect in the sense that the domestic price in each case gradually adjusts .
towards the international price.
‘The growth in the petroleum demand follows the historic pattem accurately (figure 21): Le. it
develops exponemtially until 1973. There is however a slight stagnation during this period ,
representing the stabilization of the petroleum market share. As the petroleum price increases
dramatically during the ‘70s, the demand inflects accordingly, and drops off with a substantial
delay during the ‘80s.
1i4 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA
‘The petrolewn production satisfies the demand until 1950 (fig 14). Gradually the domestic supply
is less sufficient, even if the Alaskan production comes into play. There are two reszons for this:
First of all, the import price becomes competitive in the 50's, so that the tendency to import petrol-
ew is reinforced. In spite of the import regulation, this reduces the pressure to explore domnestical-
ly. Secondly, there is a masked transition into a decline in the productivity of the investment in
exploration. This is due to the inflection of the fraction recoverable, - i.e. the declining effect of
technological progress. Even though this transition initially (1931 - 53) is very smooth, there is a
relatively dramatic fall in the net yield from exploration (figures 17, 18), partly due to the delay in
recognizing the current exploration expenditures: $o far, the productivity has been rising, and the
exploration activity has been oversized. But now, when the productivity is falling, the investnent is
systematicly too low. Gradually the effects of technological investments are deteriorating, only to
amplify the contraction. However, due to;
~ the excess exploration;
~ the large reserve that has been built up (figure 16}; and
- the relatively strong progress still characterizing production technology,
the technically recoverable reserve is inflected at only & moderate rate (figure 17). And the dramatic
fall in the yield from exploration has no effect on production. Eventually though, there is a drama-
tic decumulation of the technically recoverable reserve, and this reserve becomes an effective con-
straint on production, so that the demand can no longer be satisfied. Thus additions to the identified
reserve are no longer sufficient to maintain the required production.
‘When it eventually turns out that the production does not yield revenues, sufficient to allow the
technological progress to keep up with the depletion, then the productivity of the investments in
exploration declines. This calls for a higher price to cover the increased expenditures, and possibly
counteract the technological stsgnation. Incidentally this is exactly what happens: At this point of
time , the international price is increased, the domestic price is allowed to follow, and the high
market price stimulates investnents and revitelizes the exploration. The large invrestments are justi-
fied slso because the petrolevrn demand responds very slowly to the change in price. This creams a
shortage, previously covered by imports. But now as the international price goes up, there is an in-
centive to discontinue this import (figure 19). It finally tums out, that in spite of a boost in Alaskan
production, the domestic supply remains insufficient and one may recognize that a. contraction is in
progress. There are several factors which cause the massive investment in exploration not to pro-
vide the required reserve, and therefore may explain the subsequent contraction:
The investment is not large enough, because the exploration expenditures are under-estimated. The
effect of the investment is delayed and distributed over time. The productivity of the investment is
deteriorating exponentially along with the exploration, - an effect which is no longer adequately
matched by the technological development: While as the technological progress so far acted to
amplify the expansion of exploration and production, through positive feedback processes, it now
promotes the contraction, through the same kind of processes: The declining rate of production,
causes the revenves to decrease, - while a3 the demand for technological investnents is still grow-
ing hyper-exponentially. (Recall the significant delay cerecterizing how demand responds to a.
change in price). Even though the price is rising significantly after 1973, so a3 to conteibute to the
revenue, the technological consequences are far from sufficient. Therefore the produrtivities are
stil declining exponentially, and effectively slowing down the technological progress even further.
THE CONTRACTION (1981 -)
Historically the contraction did not manifest itself in terms of the petrolewmn price or the demand for
petrolewn (figures 22, 23): The petroleum price was detennined internationally, aud fell suddenly
down to a 1986 level that could only be justified by the costs of the largest producers in OPEC.
Because of this development, the demand slowly recovered from the price shock of the '703, after
having fallen significantly during a couple of years. The only contraction chayacteristic within the
historic timeframe was thus the production trajectory: The production decreased substantially for
about 10 years. It then levelled slightly off for the lest couple of years (figure 22).
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 115
The model reveals what is happening: Investments in exploration and production will be made as
Jong as the petroleum price may be raised to cover the additional expenditures associated with the
exploitation, As long as the price considered required to justify investments in exploration and
production, is well below the intemstionel (import) price and the price on synthetics, such invest-
ments will be made, - and the domestic price will be adjusted accordingly. Now however, a double
price ceiling is in effect, determined by.the competing petroleum sources, - imports and synthetics
(figure 20). When approaching this ceiling, domestic investments are curtailed, exploration and
production is gredvally shut down, and the exploitation is substitued by natural petroleum imports
and later on by the production of synthetics (figure 14). With a stable international price, we would
have seen this scenario play out in the '703. But due to the price increase, a large investment in
exploration is still justified in the shadow of the rising international price. The slight recovery of the
petroleum production after 1981 (figures 16, 22) is explained only a3 a delayed respons to this
large invest- ment. As it tums out however, this investment is insufficient in the long run, and as
the exploration is extended , the productivity of this investment is declining. Consequently the net
yield is rapidly diminishing, and does not support the required production (figure 18).
Note that some of the major regulating feedback processes in the systemn are deteriorating after
1953, ani completely dissapears in 1981. This is because there are no longer regulations that
effectively preserves the relationship between the exploration and production expenditures on the
one hand and the domestic price on the other. The price is not allowed to follow the increasing
expenditures, and the demand remains high.
The substantial increase in the fraction of the reserve technologically recoverable (figure 17), con-
tribute significantly to the production, because the identified reserve that remains unrecovered at
‘this point of time, is relatively lage (figure 15). But the contribution is far from sufficient to
counteract the depletion. This technological progress hardly effects costs because the exploration
expenditures are very dominating until the end of the petroleum life cyle. Gradually also the
marginal effect of this progress diminishes.
Predictions must be based upon two assumptions; one about the future economic growth and one
about the price development on the intemational petroleum market. The base case growth in GNP
exhibihed here, is the middle economic growth applied by the U.S. Department of Energy (BIA,
1985). (The model contains the extreme DOE cases as well). After 1995, linear extrapolations of
the DOE projections are applied. In the base case, the international price is assumed to reach $ 20
by 1990 (figure 20). From then on, the price is assumed to increase linearly by $ 10 per decade.
‘The tennination of the petroleum life cycle is characterized by a rapidly growing difference between
the unit exploration expenditure , and the petroletim price. The incentive to invest in exploration is
thereby eroding. The reserve is still being depleted (figure 15), until the unit production cost has
increased beyond the petroleum price, so 43 10 no longer justify production. Imports are growing
repidly during this period (figures 14,19), and one may find it realistic to assume 4 more significant
increase in the international petroleum price in response to the massive U.S. import pressure. In
that case the U.S. natural petroleum demand will be more moderate, and there will be a stonger
incentive for the domestic production of both naturel and synthetic petroleum. The model runs until
2050, but by 2020 most of the dynamics are over.
APPLICATIONS OF DEL
The most straightforward use of this model is as a tool for projecting the characteristics of the
petroleum life cycle into the future. This may be done under different scenarios with respect to the
development of the GNP, the international petrolewn price, and the price on synthetic petroleun.
The pupose of such projections would be to understand how the petroleum life cycle terminates
under different environmental circumstances. In particular, it will show how rapidly USA devel-
ops a dependency on import and synthetic production, and how large part of the total resource that
is economically recoverable. By changing the investment module of the model, the effects of differ-
116 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA
ent investment policies may be evaluated, e.g. to limit the curentexplovation end production, and
yield to the import pressure in the short xm, in order to maintain a recoverable reserve (correspond-
ing to a strategic reserve) in an attempt to smooth the long nm consequences of future import con-
straints. By changing the price module, the effects of altemative pricing policies could be tried out.
Such a policy.could include a taxation or import fees intended to stimulate energy conservation, and.
build up the production capasity for synthetic petrolewn.
Note that the model is based upon the assumption that there is a finite total petolewn resource. The
volume of this resource is continuously subject to estimation. So is the economically recoverable
part of this resource. The model suggests a way to estimate the resource: Provided one by-and-
Jorge can agree upon the other assumptions represented by the model, it may be possible to identify
arelatively nanow range of values within which we find the actual total volume of the resource.
Moreover, the model may be used to synthetically test out alternative techniques applied to acquire
‘the information neccessary to implement various policies. It was indicated that the volume of the
petrolewn resource is subject to estimation. Estimation techniques actually applied , can be de-
scribed by a set of formal models, esch of them compatible with the one discussed in this paper. In
‘that way we can carry out synthetic data experiments, in which the "real data ", acquired from the
petroleum life-cycle model, constitu: 8 consistent basis for the resource estimation, represented by
‘the additional models. Such experiments may provide a better understanding of the dynamics of
estimates, and stimulate the discussion concerning the design and utilization of estimation tech-
niques. Experiments of this kind, focusing on the Geological Analogy Method and the Hubbert
Life Cycle Method, has been carried out at MIT by J. Sterman, G.Richardson, and P. Davidsen
(Sterman and Richardson, 1985) and (Sterman, Richardson, and Davidsen, 1987).
Both the petrolewmn mode] described in this particular paper, and generalizations made to encom-
pass other depleteble resources , may tum out to be a promising tool for teaching resource manage-
ment: Such generic models will provide an understanding of how the life cycle dynamics of a
resource is related to the physical, technological ani economical characteristics of the underlying
feedback structure. Models of this kind may be used to stwly the particular behavioral modes
charecterizing the life cycle, and transitions between these modes due to shifts in the feedback loop
dominance. Based on this understanding, appropriate policies can then be suggested and evaluated
through synthetic experiments, - so also managerial information systems designed to support the
implementation of such policies.
REFERENCES
Backus, G. et. al. (1979): FOSSIL79:_Documentation. (3 vols.) DSD 166., Hanover, NH:
Resource Policy Centre, 1979.
Bell, J.A., and Senge, P.M. (1978): Methods for Enhancing Objectivity in System Dynamics
Modeling. Cambridge, The System Dynamics Group, MIT.
Coen, R. (1975): Investment Behavior, the Measurement of Depreciation, and Tax Policy.
American Economic Review, 65,59 - 74.
Choucri, N. (1961): International Energy Futures: Petroleum Prices, Power, and Payments.
Cambridge: The MIT Press.
EIA (1985): Annual Enere v Outlook 1984, with Projections to 1995. DORIEIA-0383(84).
Washington D.C.: U\S. Departmentof Energy, Energy Information Administration, -
Office of Energy Markets and End Use, January 1985
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 117
Hall, C.A., and Cleveland, C.J. (1981): Petroleum Drilling and Production in the United States:
Yield per Effort and Net Energy Analysis. Science, 211 ,4482,6 February, 576 - 579.
Hogarth, R.M. (1980): Judgement and Choice. New York. Wiley.
Hubbert, M.K. (1969): Energy Resoumes. In Committee of Resowres and Man of the Division of
Earth Sciences, National Academy of Sciences - National Research Council:
Resonices and Man; A Study and Recommendations. San Francisco: W.H. Freeman
and Company.
Hubbert, M.K. (1975): Hubberts Estimates from 1956 ty 1974 of U.S. Oil aml Gas. In
Grenon, M. (ed. : Methods and Models for Assessing Energy Resovxces. Oxford:
Pergamon Press , 370 - 83.
Morecroft, J. (1983): System Dynamics: Portraying Bounded Rationality. Omega, 11, 131 - 42.
Morecroft, J. (1985): Rationality in the Analysis of Behavioral Simulation Models. Management
Science, Vol. 31, No. 7, July 1985.
Naill, R.F, (1973): The Discovery Lifecycle of a Finite Resoue: A Case Study of U.S. Natural
Gas. In Meadows, D.L., and Meadows, D.H. (eds.) (1973): Toward Global Equili-
brum. Cambridge: The MIT Press.
Naill, R.F. (1977): Managing the Energy Transition. Cambridge: Ballinger, 1977.
Nordhaus, W.D. (1973): The Allocation of Energy Resources. Brookings Papers on Economic
Activity, vol 3, 529 - 570.
Simon, H. (1947): Administrative Behavior. New York: MacMillan.
Simon, H. (1957): Models of Man. New York: Wiley.
Simon, H. (1979): Rational Decisionmaking in Business Organizations. American Economic
Review, 69, 493 - 513.
Simon, H. (1982): Models of Bounded Rationality. (2. vols.) Cambridge, The MIT Press.
Sterman, J.D. (1981): The Energy Transition and the Economy: A System Dynamics Approach.
Cambridge: Ph. D. Dissertation, MIT.
Stenman, J.D., ani Richardson, G.P. (1985): An Experiment to Evaluate Methods for Estimating
the Fossil Fuel Resources. _ Journal of Forecasting, 4, 197 - 226.
Swennan, J.D., Richardson, G-P., and Davidsen, P.1. (1987): The Dynamics of Petroleum
Resource Estimates in the United States; Results of a Synthetic Experiment. The
Intemational System Dynamics Conference, Shanghai, 1987.
USGS (1976): Principles of the Mineral Resource Classification System of the U.S. Burean of
Mines and U.S. Geological Survey. USGS Bulletin 1450 - A, Washington, D.C.
118 THE 1987 INTERNATIONAL CONFERENCE OF .THE SYSTEM DYNAMICS SOCITY. CHINA
Unoiscovenn
INVESTMENTS; baie
DEMAND, IMPORT, AND
SUBSTITUTION “+ IN EXPLORATION:
“avenoouctION
EXPLORATIONS DISCOVERY,
ATE
z
EXPLORATION: PRODUCTION:
DISCOVERY RATE PRODUCTION RATE IDENTIFIED
_UNDENTPIEO RESERVE “DENTIFIEO RESERVE pasa
+ TECHWCALLY DISCOVERABLE “ TEGIBUCALLY RECOVERABLE
‘nesounce esenve
sRopucriny oF nvesTuENT -<Proouctiy or nveST¥ENT
THEXPLORATION renoovctiont
PnooucTion
pnonueTion (RECOVERY,
RATE
TECHNOLOGY:
FRACTION OF RESOURCE REVENUE, PRICE, AND
‘DISCOVERABLE cost cuMuLATIVE
“egeovewnate Pnopucrion
Figure 1: The sectors of the model, Figure 2: Stocks and flows.
TOTAL RESOURCE:
Teaincay
wnscoverve
FeSONTE
EMERG
woscoeen
TECUINCALY ORE
E
FESCUE
TOTAL RESOURCES
TECHNICALLY JOENTIFIED
Ea UNDISCOVERED
DeranaTaATeO wirorartieat | _ertevianive
a weasunco [| moicareo | rennco | ‘orstuer |" aunts
peNTFED 7 | erst
TECHNICALLY FESEIVE, TECINICALLY 3
REOOERE DsooMENA 3
PESEINE FESO, z RESERVE
FeManins 8 F A
i r f ee =
ecinvay el: a\8
reooewie g|3 Eiri
cum aTve FESENE 2}: [t!
He ig:
8r] 3/2:
Bie Biz:
313 ary
PROOUCTION ele zyet
2
: 1 L 1 1
INCREASING DEGREE OF
‘GEOLOGIC ASSURANCE
‘ usGs, 1976).
Figure 3: A resource classification, Figure 4: The McKelvey Box (Source:
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 119
coworen
SnoMeN
vcanienry
SHOWIN
cone)
a ce Sy
pronucnviry aes
OF INVESIMENT owe
IMEXPLORATION
TECHNICALLY
DISCOVERABLE ee
RESOURCE La,
Bonen
Figure 5: The physics of discovery.
rere
ne
seca
sate
eens
ae a
vexae
Some
poisutinn
petra “eon
INEXPILORATION |
inci BSRE wnosats
cecal wnscovee
eran
cre
Figure 7: Investments in exploration
and its technology added,
cawnron
PRODucTIN
INVESTMENT
a)
OF INVESTMENT
Ww = [ ‘a
TECHNICALLY ines [-
RECOVERABLE perallses
‘ oe 5
Figure 6: The physics of recovery.
ro
a POTENT
ehooverty ‘Y
‘OF INVESTMENT PH
|PROBUCTION ATE
TECHNICALLY foan.
ReCOVERARE
(© vecenvenenana — ERE |
wa NT | pe
recone
we. |_——"
mosscoveneo
‘nesoUNCE DISCOVERY RATE.
Figure 8: Investments in production
and its technology added.,
PERO
=
cc eeN
or nvestucnr lim,
or uwvesiuent fooucnon
WexHLORATION a over
+
TecimcaLLy
oiscovenaete
TEBOURCE netanuna FeOOe AE oh
ae, ((P esenve rouse aeceve 4],
Fractay +
roar =a raps
unorscovene
nesounice scover Rare
Figure 10: The demand for production
Figure 9: The demand for exploration
added.
or
<P syvesiventey
BcrAneR TEC neLOGY
Frvcron Frkorcn
DSCOVET ELE Foe
Figure 11: Shift in technology investments.
PRICE ON
memuaTin. syns
‘upon PRIcE
‘uur exeLORATION
‘orcas \
ean
igs
cL nia
rou:
BIvESTHENT wy PRODUCTION
PETROLEUM
SUPPLY Permoteum
BEMARO
Figure 12: Factors that determine the
Petroleum price.
added.
iurenniarionat, price oF | Grp
(MPORT PRICE SYNTHETIC
untr costs oF PETROLEUM
EXPLORATION/ PRODUCTION +> Price
DEMAND FOR
‘EXPLORATION / PRODUCTION
+
INVESTMENT itt
EXPLORATION) PRODUCTION
MEOF
PEGE piscoveny /PRODUCTION
TECNICALLY DISCOVERARLE/
‘RECOVERABLE RESERVE REMAINING
PRODUCTIVITY OF INVESTMENT IN. +
EXPLORATION / PRODUCTION
Figure 13: The influence of the petroleum
price.
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 121
——_ == NPDHND
— — - NHI
9000. 26
BARRELS
PER YEAR 3
6000, 06 $s othe:
BETS
wt
4000, 26 as
2000.26 Lf ma
LY see =~
PROPUCTION IN|U.S
a x
LOWER: 48 STATES ae
1990, 1920, 1940, 1 1980, 20d, 2028;
Figure 14: Petroleum Production (PR) and the contributions from Imports,
Production In Alaska, and the production of Synthetic Petroleun.
Giink aeneeeseesTDRR
ed —e
some
800.29,
BARRELS
600.29
[ee UR
ial
oN
498,29
4 _ ) IR
ge ee me ae — oe ae oe
200, xX : cues)
LL
fs
a :
1900, 1920, 140, 1969. 1980, 2000, = 202,
: TIME
Figure 15: Cumilative Production (CUMPR), Undiscovered Resource (UR),
Identified Unrecovered Reserve (IR), Technically Discover-
able Resource Remaining (TDRR), and Technically Recoverable
Reserve Remaining (TRRR).
122 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA
BARRELS
PER YEAR
6020, 26)
4900, per
a
Pi
2000.0 CL Ze As.
At i
—— cll
1900, 1928, 1049, ai 1980, 2008, 2028,
Figure 16: Petroleum Production in the lower 48 states(PR), Potential
Production from Reserves (PPR), and Potential Production
(investments taken into consideration) (PP).
PQ 1.) === =~ TDR, 20829)
L—— AERO) eeeeeeeees TRRRCG., 289, 29)
200,84
--/
BARRELS a
= ne
5 i
wo = I TT
Leo ER 7
100.33 as
A FR salle
vA \ qo
rl r Lew
sahedL, eal a7 '
—— a ve wa "RRR
a Se een. bie ree
Q eens” Ti eee =
a 7
130d, 1928, 1348, 1969, 1988, 2008, 2020,
TIME
Figure 17: Fraction Discoverable (FD), Fraction Recoverable(FR), Tech-
nically Discoverable Resource Remaining (TDRR), and Techni-
cally Recoverable Reserve Remaining (TRRR).
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 123
eS AIR
——-PR
20,29
BARRELS
PER YEAR
13,89
10.28] AIR a
A \
lL 4 \
5060.2 / at —
PPR, ny
J _ a or ae
a ee “ale PP,
a ore
1900, 1928, 1949, 1960, 1989, 200d, = 2023,
TIME
Figure 18: Petroleum Production (PR), Potential Production from Reser-
ves (PPR), and Additions to Identified Reserves (AIR).
THPORT — — -IIAPRT
6000.26
BARRELS
PER YEAR
4000.26
2000.06 * 7
-2000,06
, 1908, 1928, 1943, win 1980, 2000, 2020,
Figure 19: Imports (IMPORT) and (Price-) Indicated Import (IIMPRT).
124 THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA
mE ee SPRICE
Wo TPRICE
DOLLARS
(1982)
6a, 2 i
20,
A 2
PN TA eel | #
apRice >
a :
1900, 1928, 1949, 1268, 1338, 2008, 2029,
TIME
Figure 20: Petroleum Price (PRICE); Import Price (IPRICE), and Price
on Synthetic Petroleum (SPRICE).
NPOHNDH — — -NPDKND
9000, ¢6
BARRELS
PER YEAR
6000, 26 ZA
,
|/PDMND /
4000. 06 UE _
ym
200026 "4
a
1870, 1838, 1310, 1938, 1980, 1970, 1986,
TIME
Figure 21: Historic Natural Petroleum Demand (NPDMNDH) versus Simulated
Demand (NPDMND) .
THE 1987 INTERNATIONAL CONFERENCE OF THE SYSTEM DYNAMICS SOCITY. CHINA 125
—— PRHIST a
8000, 26, o
BARRELS
PER YEAR
6000, 06;
4000.26 ote
Xs
a UN
2900, 26; C7 a
a
1970, 1898, 190, 1939, 1950, 1970, 1986.
TIME
Figure 22: Historic Petroleum Production (PRHIST), versus Simulated
Production (PR).
PRCRDPP — — ~PRODPP
DOLLARS
(1982)
ue
ti i
N\A
area is y
/’ N \ vin (OQ.
wy We
PRERDPP
ig.
‘ 1870, 1898, 191d, 1998, 1950, 1970, 1986,
TIKE
Figure 23: Historic Real (1982 USD) Price on Domestically Produced
Petroleum (PRCRDPP) versus Simulated Price (PRCDPP).
