2

LIMITATIONS OF RESOURCE ASSESSMENTS

It is essential to recognize that estimates of undiscovered oil and gas resources are just that: estimates. They are an attempt to quantify something that cannot be accurately known until the resource has been essentially depleted. For that reason, resource estimates should be viewed as assessed at a point in time based on whatever data, information and methodology were available at that time. Resource estimates therefore are subject to continuing revision as undiscovered resources are converted to reserves and as improvements in data and assessment methods occur.

Historically, estimates of the quantities of undiscovered oil and gas resources expected to exist within a region or the nation have been prepared for a variety of purposes using several different methods. To make effective use of such estimates, or to compare them with others, one must develop an understanding of how and why they were prepared; the extent and reliability of the data upon which they are based; the expertise of the assessors; the implications and limitations of the methodology used; and the nature of any geographic, economic, technologic, or time limitations and assumptions that may apply. It is equally important that those who prepare estimates provide documentation adequate to allow the users to evaluate the issues just described. The purpose of this chapter is to examine, in general terms, some of these issues and how they may impact on the credibility and usefulness of resource estimates.

ASSESSMENT OBJECTIVES

Resource estimates serve many purposes. They may be prepared simply to inventory various energy commodities to evaluate future supply options. They



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures 2 LIMITATIONS OF RESOURCE ASSESSMENTS It is essential to recognize that estimates of undiscovered oil and gas resources are just that: estimates. They are an attempt to quantify something that cannot be accurately known until the resource has been essentially depleted. For that reason, resource estimates should be viewed as assessed at a point in time based on whatever data, information and methodology were available at that time. Resource estimates therefore are subject to continuing revision as undiscovered resources are converted to reserves and as improvements in data and assessment methods occur. Historically, estimates of the quantities of undiscovered oil and gas resources expected to exist within a region or the nation have been prepared for a variety of purposes using several different methods. To make effective use of such estimates, or to compare them with others, one must develop an understanding of how and why they were prepared; the extent and reliability of the data upon which they are based; the expertise of the assessors; the implications and limitations of the methodology used; and the nature of any geographic, economic, technologic, or time limitations and assumptions that may apply. It is equally important that those who prepare estimates provide documentation adequate to allow the users to evaluate the issues just described. The purpose of this chapter is to examine, in general terms, some of these issues and how they may impact on the credibility and usefulness of resource estimates. ASSESSMENT OBJECTIVES Resource estimates serve many purposes. They may be prepared simply to inventory various energy commodities to evaluate future supply options. They

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures provide data essential for appraising state or federal lands prior to leasing or sale. They may be undertaken to compare the relative merits of oil and gas development versus other uses for land. Large corporations and financial institutions use resource estimates for long-term planning and analysis of investment options. Industry groups, like the American Gas Association, the American Petroleum Institute, and the Potential Gas Committee (PGC), use them as guides to the future health of their industry. Increasingly, governments and the public are looking for resource estimates to provide objective statements of how much oil and gas will be available for future domestic consumption. It is important to understand the purpose underlying any estimate, because it controls to some extent several factors that influence the assessment's outcome: the methods used, the geographic scope, the economic assumptions, the skills and biases of analysts, and the conclusions about the timing of resource availability. For some purposes, methods such as the simple extrapolation of historic resource discovery rates without regard for economic or technological change may suffice. Other uses of resource estimates may require methods capable of estimating the size ranges of individual accumulations and their reservoir properties to analyze future supply economics. Certainly if the objective of a national resource estimate is to determine if sufficient oil and gas will be available to maintain an acceptable standard of living and a viable industrial complex, then one must be concerned about levels of accuracy and the adequacy of data and analysis. One must be assured that all potential oil and gas sources have been considered. Reactions of citizens and governmental bodies, and resulting public policy, will be quite different if there is a perception of future domestic abundance rather than a perception of future domestic shortage; consequently, the need for accuracy is paramount in national resource estimates designed to help determine public policy. UNCERTAINTY IN ASSESSMENTS Uncertainty is an integral part of resource estimates. Almost every component of the assessment process has associated uncertainty, and the aggregate level of uncertainty for the final resource estimates can be large. Major uncertainties arise from limitations in the data base and methodology, difficulties in projecting the course of recovery technology, and the sensitivity of the economic analysis to changes in energy prices and production costs.

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures In general, uncertainties in estimates of undiscovered oil and natural gas are greatest for areas that have had little or no past exploratory effort. For areas that have been extensively explored and are in a mature development stage, many of the unknowns have been eliminated and future resources can be evaluated with much more confidence. Even in some mature producing areas, however, uncertainty remains about the potential oil and gas supply at greater drilling depths. Uncertainty also pervades projections of whether potential reservoirs have been unrecognized or bypassed in past drilling. Similarly, where resource estimates are based on analogue comparisons between maturely explored areas and unexplored areas, uncertainty is introduced because each area or basin has unique characteristics. Although our fundamental knowledge of the origin, migration, and entrapment of oil and gas has advanced markedly during the past 30 years, the fact that incremental scientific advances are still being made leads to additional uncertainty in resource estimation, especially in frontier areas or at great drilling depths. In other words, new knowledge may lead to increases or decreases in estimates of undiscovered resources, but generally leads to a reduction of uncertainty. Discovery is only the first step in crude oil and natural gas resource development. The present state of technical knowledge in reservoir geology and petroleum engineering, as well as existing regulations, determine the spacing, completion, and production methods of development wells (i.e., petroleum producing wells drilled after the discovery well). As engineering and geologic knowledge increase, our ability to withdraw larger increments of oil and gas from existing fields is enhanced. Thus, the sizes of fields in terms of ultimately producible barrels of oil or cubic feet of gas increase with time. Uncertainty as to ultimate sizes of discovered fields leads to uncertainty in estimates of size distributions of undiscovered fields in areas with analogous reservoir characteristics and geologic histories. Scientists can estimate the quantity of technically recoverable undiscovered oil and gas based on the present state of geological and engineering knowledge, modified by a consideration of future technological advancement. However, the percentage of that quantity that may actually be discovered and produced is an economic question. Uncertainties about future well-head crude oil and natural gas prices and costs of exploration and development adversely affect all resource estimates. In short, uncertainties embodied in economic assumptions lead to significant uncertainties in estimates of economically recoverable resources and account for some of the large differences among estimators.

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures There are no foolproof, completely mechanical methods for estimating undiscovered resources. Because all methods contain elements of subjective judgement or expert opinion, the degree of uncertainty is affected by the level of expertise of the personnel doing the estimating, the time devoted to the estimates, the methods by which the estimates are tested by peer review, and the level of enthusiasm for the appraisal. ASSESSMENT BOUNDARIES In general, policy concerns about oil and gas resources are actually concerns about future domestic production rates. Policymakers tend to view resource estimates as indicators of production capability. However, the results of resource assessments are not often presented in a form that recognizes this concern with future production. Instead, the results commonly represent only a piece of the resource base. The USGS/MMS assessment evaluated in this report covers conventional undiscovered recoverable oil and gas—only a part of the total resource supply that will contribute to future production. To the uninformed user of this assessment, it might seem that what the assessment includes is straightforward: undiscovered petroleum recoverable with existing technology. Yet, in this and other resource assessments, the distinctions between “conventional” and “unconventional,” “recoverable” and “unrecoverable,” “discovered” and “undiscovered” are hazy. To interpret the results of an assessment correctly, the user of the assessment must understand where the assessment' s boundaries are drawn—what is and what is not included in the assessment. Discovered/Undiscovered Boundary To some, the difference between undiscovered and discovered petroleum might seem obvious. The term “undiscovered” implies a clear distinction between resources we have found, identified, and measured and those we have not. Yet, in some cases, deciding what to classify as discovered is a judgment call that depends on an analyst's interpretation of oil-industry definitions. Resource appraisers divide the nation's remaining oil and gas reserves into four categories, familiar to those in the petroleum industry:

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures Measured reserves are contained in known reservoirs, have been quantified by engineering studies, and are recoverable under existing economic and operating conditions. Indicated reserves are reserves from known reservoirs for which extraction depends upon improved recovery techniques. Inferred reserves are as-yet undocumented resources that analysts expect may be added to existing petroleum fields from extensions and additional development. Undiscovered resources are those outside of known petroleum fields that analysts postulate to exist based on broad geologic knowledge. Figure 2.1 diagrams the distinctions between these types of reserves. On the figure, level of certainty that the reserves exist increases from right to left. The distinction between inferred reserves and undiscovered resources may confuse users of assessments. Depending on how an assessor interprets the above definitions, a postulated but unknown reservoir that a layman might believe is “undiscovered” can be defined as “inferred” and therefore categorized as “discovered.” Conventional/Unconventional Boundary The term “conventional” implies that a common method of resource extraction or a well-identified set of physical characteristics clearly separates “conventional” from “unconventional ” petroleum. This is at best an oversimplification. The boundaries between conventional and unconventional resources have always been somewhat indistinct and have become increasingly so over the past several years. This complicates comparisons between alternative resource assessments and interpretation of individual assessments. Problems with maintaining strict boundaries include: Relatively large quantities of natural gas from sources traditionally labelled “unconventional” are being produced in some areas. Examples of “unconventional” resources whose production is becoming more common are coal-seam methane and gas from low-permeability sandstone reservoirs (called “tight gas”) and fractured shale reservoirs. Technologies used to produce unconventional gas, especially fracturing, are now being widely used in formations that are considered “conventional. ”

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures FIGURE 2.1 A diagram showing how petroleum resources are classified. The area within the heavy frame on the upper right represents the undiscovered recoverable resources estimated in the 1989 DOI assessment. The hachured area within the heavy frame indicates undiscovered resources that are estimated to be economically recoverable. Source: U.S. Department of the Interior, 1989. Reservoir permeability, which is the defining characteristic for tight-gas reservoirs, can vary widely within individual formations; many formations combine permeability characteristics on both sides of any defined conventional/unconventional boundary. Similarly, gas withdrawn from coal seams may contain gas from both the seam and the adjacent sedimentary strata—that is, a combination of conventional and so-called unconventional gas. Available data sources cannot distinguish between conventional and unconventional gas in these circumstances. In the past, analysts placed resources in the “unconventional” category for two reasons: to designate resources that were considered a potential but not a current source of production (except perhaps in small quantities), and to identify resources that required production methods significantly different from those

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures normally used. Thus, as production of at least a portion of the unconventional resources has grown, and as the technology for this production has become widely used in the industry and thus “conventional,” it makes sense to move some of the unconventional resources out of this category and into the category of conventional. Allowing the boundaries between conventional and unconventional resources to be frozen according to obsolete conventions and practices fails to recognize important technological advances. Recoverable/Unrecoverable Boundary Most assessments use two criteria to define the boundary between recoverable and unrecoverable resources: technology and economics. Technically recoverable resources are resources for which technology exists that can locate and produce the oil and gas and transport it to market. Economically recoverable resources are those that a profit-driven industry would seek to exploit. (The USGS/MMS assessment gives two estimates of undiscovered oil and gas reserves: one that includes resources recoverable with conventional technology and one that includes only the recoverable resources that yield an 8 percent real profit.) The problem with categorizing resources as technically or economically recoverable is that changing assumptions and changing times can drastically alter these criteria and hence the assessment results. Technically Recoverable Resources Resource assessments should, but rarely do, begin with a calculation of in-place resources: the total volume of petroleum trapped within the play being assessed, without consideration of whether or not the petroleum is extractable. Instead, assessments typically limit their evaluation to resources recoverable with conventional technology. Thus, when assessors consider what resource volume exists in a petroleum field, they weight their considerations with a judgment about how much of the resource is recoverable. In part, this is because until recently, obtaining data on in-place resources was difficult, as this chapter will discuss later. The distinction between in-place resources and recoverable resources arises because a substantial part of the earth's total petroleum content is either dispersed at very low concentrations throughout the crust or is present in forms that cannot be extracted except through methods that would cost more than the

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures petroleum is worth. An additional portion of the total petroleum resource base cannot be recovered because available production technology cannot extract all of the in-place crude oil and natural gas even when these resources are present in commercial concentrations. This is particularly important for crude oil; using current practices, only an average of one-third of the total in-place crude oil in a reservoir can be recovered by primary and secondary recovery processes. The remaining two-thirds of the resource may not be recoverable at the present time, but it does remain a target for future technology. For natural gas, a higher percentage (60 to 85 percent) of in-place reservoir natural gas is recovered using current technology. The inclusion of oil and gas resources in the category of technically recoverable or technically unrecoverable is at best a snapshot in time, and, if history is to judge, one destined to be quickly rendered obsolete. The development of improved technology continually causes resources to move from the unrecoverable to the recoverable portion of the resource base. Improvements in offshore drilling technology have allowed drilling in deeper waters and more hostile conditions, opening up new territories to development. As a result, succeeding generations of natural gas resource assessments have steadily moved their boundaries to deeper and deeper waters. Other advances in offshore technology have lowered costs sufficiently to allow the exploitation of fields previously considered too small to develop profitably; thus, technical advances may move resources across economic as well as technical boundaries. Refinements in seismic techniques have allowed exploitation of the Western Overthrust Belt, which has very complex natural gas-bearing formations. (Here, there may be arguments about whether the boundary crossed was technical or economic, since intensive conventional drilling might have revealed the gas fields in this area.) Progress in horizontal drilling technology has opened production zones that otherwise would be uneconomic. Improvements in drill bit metallurgy, downhole drill control, and drilling mud systems have allowed exploration of deep natural gas deposits unreachable by earlier, “conventional” drilling technology. The language describing technology assumptions in current resource assessments is sufficiently vague to inspire doubts about just where the technological boundaries lie. The Potential Gas Committee (PGC) assumes the use of “current or foreseeable technology.” The DOI defines its technological boundary by exclusion: it excludes all resources except those that can be produced by “natural pressure, pumping, or injection of water or gas.” It does not

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures mention exploration technology, even though an inability to locate resource deposits in complex formations can prevent recovery. Our concerns about precisely where the technological boundaries of an assessment center on the following issues: How does the assessment treat resources that theoretically could be recovered with conventional technology used at an unusual degree of intensity (for example, drilling at a closer-than-standard spacing) but that are unlikely to be economically recovered without development of new technology? Many technologies that are “commercial” remain the province of a small portion of the potential users or are restricted to a few geological areas. Does the assessment assume that these technologies will be in use universally? Just what does “the foreseeable future” mean? In the absence of specific assumptions about exploration technology, does the assessment include all oil and gas fields containing economically recoverable resources—regardless of whether the technology for finding them is economical or available? With the exception of a limited effort by the PGC, no assessors have attempted to identify the extent to which improved technology has caused resource estimates to increase over the years. Without the availability of such information, it is difficult to quantify the effects of technological improvement on total recoverable resource volumes. Consequently, it is difficult to predict credibly how future assessments are likely to change with technological improvements. However, we can identify the advances that are most likely to increase recoverable resource volumes in the near future. These include: extension of horizontal drilling to common use, which will boost accessible resource volumes in thin pay zones and in highly fractured and compartmentalized reservoirs; advances in chemical-enhanced oil recovery methods, which will raise recovery efficiencies in fields undergoing tertiary recovery; progress in reservoir characterization and modelling, which should enhance the potential for using infield drilling to add to reserve growth; and improvements in subsea completions, which will allow the development of smaller fields and fields farther offshore and in deeper water than was previously possible.

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures Economically Recoverable Resources Of the technically recoverable resource, assessors wish to eliminate resources that exist in volumes too small, in waters too deep, in strata too far underground, in places too remote, or in physical conditions too difficult to allow for profitable extraction and shipment. Users of the assessment must recognize, however, that the location of the economic boundary will depend on both the economic assumptions chosen and economic conditions at the time of the assessment. Economic boundaries are introduced into resource assessments in one of three ways: (1) through quantitative analysis using explicit assumptions about resource prices, economic conditions, and available technology; (2) through quantitative analysis using implicit economic assumptions like minimum economic field size (the smallest size field assessors determine is profitable to develop) and water depth limits; or (3) through subjective, qualitative assumptions about economic relationships. An example of (3) is the PGC's definition of economic boundaries: “only the natural gas resource which can be discovered and produced using current or foreseeable technology and under the condition that the price/cost ratio will be favorable. ” With (1) and (2), assessors attempt to define precise limits of economic attractiveness. They incorporate assumptions that may include an explicit price scenario that varies over time, a statement of the tax regime under which resource exploration and production take place, and a minimum acceptable rate of return. The MMS and USGS used this latter approach in their assessment. Defining the economic boundaries of a resource base is a controversial issue in resource assessment, primarily because the economic attractiveness of resources is volatile over time and depends on a complex set of variables, including the state of technology. Furthermore, economic analysis undertaken during the past few years has been especially problematic for oil and gas resource assessment. The events of the past 15 years have demonstrated the instability of both costs and prices: the rapid “boom” in oil-industry activity from the late 1970s until about 1981, followed by the gradual decline from 1981 until 1985, followed by the plunge in 1985 and 1986 in both world oil prices and activity levels. During the oil price increases of the 1970s and early 1980s, drilling and service costs rose sharply and industry efficiency fell. As a consequence, the actual price-driven expansion of the recoverable resource base was considerably below what might have occurred had costs and efficiencies remained constant. Similarly, although the gradual and then sharp oil price drops of the 1980s

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures pushed many prospects out of the range of economic viability, the effect was smaller than expected. Large declines in drilling and field-service costs have moderated the negative impact of the price drops, limiting the declines in operator profitability. In addition, the economic pressure that oil and gas operators experienced during the past few years stimulated major efforts at cost reduction and increased efficiency through better selection of prospects. These efforts have further alleviated the effects of the price drop. Thus, it has become common to learn about prospects being developed at $18 per barrel of oil that a few years ago were believed to require $25 per barrel or more to develop profitably. The positive relationship between oil and gas prices and drilling and service costs, described above, implies that resource-base boundaries will be less sensitive to resource prices than would be expected from economic analyses assuming constant costs. Nevertheless, changes in resource prices or in other economic conditions—such as tax rates or allowed tax credits or income deductions—can add to or subtract from the economically recoverable resource. Users of any resource assessment should be aware of the nature of the assumptions that define the resource boundaries. In addition to price fluctuations, another factor that strongly affects the economics of oil and gas resource development, and thus the volume of oil and gas within an assessment's “economically recoverable” boundary, is the available infrastructure: pipelines, ports, gas processing plants, and so forth. Where a strong infrastructure already exists, resources that would otherwise be uneconomic—those located in small fields or in difficult and expensive operating environments—often may be developed profitably. Most areas within the onshore lower 48 states have a strong existing infrastructure. In more remote areas, however, the decision to develop must be predicated on having to invest heavily in infrastructure. For example, in many areas of Alaska's North Slope, development of future discoveries cannot proceed without building a pipeline to the Trans Alaskan Pipeline System and constructing extensive port and airfield facilities. For such an investment to be profitable, explorers must find large fields, with oil volumes of several hundred million barrels. In addition, natural gas is excluded from the economically recoverable resource base, because no pipeline exists to transport the natural gas to markets in the lower 48 states. Assessment users should understand that once infrastructure is constructed, future assessments can change drastically. For example, if a pipeline were built from Prudhoe Bay to the Arctic National Wildlife Refuge, future resource

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures BOX 2.1 How to Define a Play In defining a play as “a group of geologically similar prospects having basically the same source-reservoir-trap controls of oil and gas,” White stressed the importance of geologic commonality in play definition (White, 1980). Achieving geologic commonality is essential for insuring that each set of reservoirs and prospects being evaluated is as homogeneous as possible. Yet, in practice, achieving commonality is not a straightforward exercise. There is no single operational formula for play definition that fits all geologic settings. In the great majority of cases, plays are limited to a single formation, because each formation is associated with a distinct set of reservoir characteristics. Yet in some areas, notably California and some Rocky Mountain basins, a single play may encompass several producing formations, the play being essentially defined by a structural trend. Within a single formation, depositional system can be a key parameter in play definition, because differences among types of depositional systems are associated with differences in reservoir size distributions and patterns of reservoir location. Differences in petroleum sources, in thermal maturity within a source, and in migration history as they affect petroleum type and characteristics can also be important factors in determining play boundaries. applied the widely recognized concept that petroleum pool or field sizes appear to be lognormally distributed in nature (Kaufman, 1965). Most importantly, the new methods made extensive use of subjective probability. (Roy et al., 1975; Roy, 1979; Energy Mines and Resources, Canada, 1977; White and Gehman, 1979).

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures Subjective Probability Methods Central to the new subjective probability methods was the solution of some variant of the standard engineering equation for calculating the reserves in an individual pool or field, but applied to all of the prospects in a play. In this process, a frequency distribution appropriate for all prospects represents each variable in the equation (net pay, area, porosity, hydrocarbon saturation, etc.). These distributions incorporate all the observed values that result from discovery, supplemented with the assessment team's subjective judgment. The distributions are inserted in the standard engineering equation in place of single number values for the variables. Where the equation requires, analysts multiply the distributions together with an appropriate mathematical procedure, such as Monte Carlo simulation. The equation then produces a conditional pool-size distribution: a curve showing the possible sizes of petroleum pools plotted against their frequency of occurrence. The pool-size distribution is called “conditional” (or, alternatively, “unrisked”) because it assumes the condition that the play contains some minimum amount of petroleum. In other words, analysts consider the possible size attributes of petroleum deposits separately from the question of whether the play contains any petroleum. One can view the probability that a play contains petroleum as analogous to the chance of success of a wildcat exploratory well. Determining this probability requires knowledge of the geologic factors—a source rock, a reservoir rock, a trap, and so forth—necessary for petroleum accumulation. Based on geologic evidence, analysts assess a marginal probability for each such geologic factor. Then, they multiply the marginal probabilities together (a calculation that requires the sometimes questionable assumption that the factors are statistically independent) to obtain the chance that the play contains petroleum. By combining the exploratory well success probability, the conditional pool-size distribution, and a second distribution representing the number of prospects in the play, analysts produce a probability curve showing the total petroleum volume in the play. Subjective probability methods gained favor with assessors in part because they can be geology-based and because they provide a relatively simple means of reflecting uncertainties associated with the variables that describe pool size. A problem with such methods, however, is that the Monte Carlo combination of size variables is legitimate only if each variable is functionally independent. Empirical evidence has shown that this is not always the case. There may be

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures correlations between area and net pay; porosity and depth; and other sets of variables. More recently, analysts have altered programs to address possible dependence between size variables. A variation on the subjective probability approach forms the basis for part of the MMS's PRESTO (Probabilistic Resource Estimates—Offshore) model: a computer program for simulating exploratory drilling. The PRESTO program applies subjective probability to assess individual prospects which can then be summed into plays. The subjective probability methods described above can be shown to be statistically valid (Lee and Wang, 1983a,b), but commonly serious problems arise in their implementation. One problem is that analysts typically have little or no training in assessing uncertainties or in evaluating probabilistic dependencies between variables. A second problem is that an assessment team commonly reaches consensus at the expense of capturing the total spread of opinion regarding uncertainty. Assessments disguise the opinions of individual assessors by combining these opinions and publishing only a single, group probability distribution. A third problem is the absence of protocols to ensure consistency in assessment methods. Lacking such protocols, assessors may interpret differently even terms that seem relatively straightforward, like “maximum value” and “minimum value.” For example, one assessor may interpret the “minimum value” of the number of prospects in a play as a value such that a small but non-zero probability exists that the number of prospects is less than this “minimum.” Another assessor may interpret the same term to mean that there is zero probability that the number of prospects falls below this minimum. Assessors are generally aware of the potential problems with these subjective probability methods, but feel that there is such a large error cloud involved that many of the problems can be safely ignored. Discovery-Process Models To sidestep the problems with subjective probability methods, assessors developed a new generation of methods that centered on objective, probabilistic models of the petroleum discovery process. Such “discovery-process models” are built from assumptions that describe both physical features of petroleum deposits and fields and the manner in which they are discovered. Their principal premise is that discovery data are size biased: large deposits are more likely to be discovered early in the evolution of a play than are small fields (Kaufman et al., 1975). Thus, exploration produces a sample of the petroleum pool population

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures that is biased toward early discovery of larger pools. By understanding the nature of the bias and modelling it mathematically, Kaufman derived the parameters describing the North Sea petroleum pool population and successfully estimated resource values for North Sea plays. Lee and Wang built on Kaufman's work, employing somewhat different mathematical procedures to derive the population parameters and adding the capability to determine individual pool sizes (Lee and Wang, 1985). Assessment methods based on discovery-process models have two obvious advantages. First, the input required comes from two of the most readily available and reliable types of geological data: the sizes of discovered deposits and the order of discovery. Second, the assumptions defining the model can be tested for validity. A difficulty of the process, common to all assessment procedures, is that assessors must be careful in defining exploration plays. Faulty definitions can result in mixing of plays, which can seriously distort the model 's projections of undiscovered oil and gas, as is shown in Appendix C. In the assessment this report reviews, the USGS employed a hybrid of subjective probability methods and discovery-process models. A fundamental concept from discovery-process models is that a play 's field-size distribution (the graph showing the frequency at which each possible field-size occurs) shifts predictably with time. As exploration progresses, on average, the proportion of large deposits remaining undiscovered decreases rapidly relative to the proportion of smaller fields; the field-size distribution shifts toward a smaller percentage of large fields. Thus, the return on investment decreases as exploration progresses, as the large fields are exploited and the less profitable small fields remain. By applying this concept and creating different field-size distributions for different points in time, USGS analysts projected the rate of falloff of returns on exploratory efforts. Subjective probability entered into the USGS assessment in the way analysts created field-size distributions. Instead of applying objective discovery-process models, USGS analysts used subjective judgment and a function called a “Pareto” distribution: a probability curve that analysts can tailor to fit known field-size data. To create field-size distributions, analysts fit sizes of known oil and gas fields in each play to Pareto distributions. They “truncated” the distributions at the maximum field size and “shifted” their origins to a minimum field size; hence, the equation that describes the resulting curve is called a “Truncated Shifted Pareto” (TSP). Fitting Pareto distributions to data from discovered fields requires expert judgment. Thus, one can view the TSP function as an aid for translating expert opinion about field sizes into mathematical language.

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures To produce their final resource estimates, USGS assessors combined the TSP distributions with independently assessed distributions for the number of undiscovered fields in each play. Methods for Considering Economics As noted earlier, assessors may avoid quantitative economic analyses and the necessity for establishing explicit price and cost scenarios by defining economic boundaries qualitatively, as in the PGC's boundary of a “favorable” price/cost ratio. Problems with this approach include the difficulty of reviewing the estimates without precise boundary definitions, and the possibility—or probability—that individual assessors will apply different economic assumptions based on their own interpretations of this vague boundary condition. (At the PGC, discussions within working committees attempt to ensure that economic boundaries are applied consistently.) An advantage of this approach is its ability to recognize regional differences in production costs, infrastructure, and petroleum prices. When an assessment includes explicit, quantitative economic analysis, analysts may choose from a number of economics “scenarios”: They may use an actual oil and gas price forecast that varies over time, and assume drilling and other development costs are constant (adjusted only for inflation), as in the USGS/MMS assessment. They may present the different resource estimates for different oil prices, ranging, for example, from $15 to $30 per barrel in $5 increments. They may define precise physical boundary conditions tied implicitly to economic conditions: water depth limits, minimum economic field size, permeability limits, drilling depth limits, and so forth. Analysts have shown that assessments of economically recoverable oil and gas vary considerably with world oil price. For example, the MMS evaluated the sensitivity to price of estimated recoverable oil and gas resources in offshore federal lands for its 1987 Five-Year Outer Continental Shelf Leasing Program document. This analysis found that leasable resources on the OCS varied from

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures 9,100 million barrels of oil equivalent (Mmboe) when oil cost $17 per barrel 1 to 13,690 Mmboe when the price rose to $34 per barrel. 2 In other words, halving prices drove down recoverable resource estimates by 34 percent. The sensitivity of estimated resource volumes to oil price depends on assumptions about costs. It is quite common to assume that oil field costs are constant (in real dollars) over time, regardless of oil prices. However, as explained earlier in this chapter, research has shown that oil-field service costs increase with oil prices (Kuuskraa et al., 1987). If oil prices rise quickly, stimulating immediate increases in drilling rates, service costs will rise rapidly as well. Similarly, a rapid fall in prices, with a sharp drop in drilling and rig utilization, would drive down costs. More gradual price changes would cause less dramatic changes in costs because rig supply would have time to adjust, and shortage or surplus situations would not contribute to swings in service costs. This added complexity in economic analysis—that oil-field service costs vary with changing prices and the rate at which prices change—creates a serious dilemma for the resource analyst. It implies that an appropriate range of scenarios for a credible quantitative analysis of recoverable resources extends to predicting not only future price levels, but also how fast they are reached and how the service industry responds. On the other hand, the variability of service costs tends to damp out the sensitivity of recoverable resources volume to price, because changes in service costs eat up some of the potential profits or losses associated with higher or lower prices (Office of Technology Assessment, 1987). Given the importance of drilling and other costs in analyzing the economic recoverability of resources, and the potential for cost volatility, resource assessors should seek baseline economic assumptions that reflect long-term equilibrium costs, rather than costs that have been artificially inflated or deflated by rapid price changes. These costs would reflect the assumption that the supply and demand of oil-field and gas-field services are essentially in balance. This does not mean that such equilibrium costs will be constant over time, because they will change with technical advances. Unless the assessors can define clear trends in such costs, however, they probably should assume constant real costs over time. If it is desirable, assessors can examine the effect of changing costs using 1   In 1987 dollars, delivered at the Gulf of Mexico. 2   U.S. Department of the Interior, Minerals Management Service. Appendix F: Economic considerations in the 5-year outer continental shelf oil and gas leasing program, in 5-Year Outer Continental Shelf Leasing Program for January 1987-December 1991, draft.

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures parametric analysis. (That is, they can recalculate resource volumes at alternative costs.) Similarly, an ideal assessment would use long-term equilibrium values of oil and gas prices, avoiding values that can be sustained only for short times. Unfortunately, there is little likelihood of obtaining consensus on what these prices would be, so it makes sense for assessors to use well-known price forecasts like those from the Energy Information Administration's Energy Outlook series. Additionally, to account for the role efficiency plays in prospect development one should choose economic and cost assumptions that reflect an “optimistic” rather than an “industry average” outlook. This choice is dictated by the nature of oil and gas development. For most prospects, there are several potential developers who, as a group, may represent an industry average in terms of drilling and service costs, required rate of return, exploration success rates, and so forth. However, it only requires one company to develop a prospect; the fact that all but one of the potential developers could not develop the prospect profitably (or would not proceed with development according to their investment decision rules) is not relevant to the prospect's inclusion in the economically recoverable resource base. Thus, the required rates of return assumed in the assessment should fall near the lower bound of all rates applicable to the group of producers capable of developing the prospects. Similarly, the assumed drilling costs should not represent the average, but should approach the costs achieved by the lowest-cost driller. It is not appropriate, however, to select the absolute lower bound (or upper bound, as appropriate) for these variables, as no individual producer is likely to be both the lowest-cost and most efficient performer in every category. Instead, one should perhaps select the lower (or upper) decile or quartile value for each variable. A controversial aspect of selecting economic assumptions is whether or not to treat certain front-end costs as “sunk” (that is, as already spent and therefore not relevant to future investment decisions). Assessors may, in particular, treat leasing and exploration costs as sunk, on the basis that the critical economic/uneconomic decision often comes after exploration drilling, when a company has found a field and must decide whether or not to develop it. Under this assumption, assessors calculate minimum field sizes for assessment areas by comparing the value of the oil and gas found to the costs of developing the resources (i.e., the cost of building infrastructure and drilling development wells), without considering lease or exploratory drilling. The validity of treating leasing and exploration costs as sunk depends on negative answers to two questions:

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures Might developers choose to reject prospects in advance of exploration because they fear high exploration costs? If the field is abandoned before development commences, can developers recoup any of their exploration costs (i.e., is there any salvage value)? An affirmative answer to either question implies that leasing and exploration costs cannot be ignored in deciding whether potential resources are economic. A negative answer to the first question implies that, for all of the prospects considered, at least one company in the industry will be willing to buy a lease and conduct an exploration program. Individual geologists and companies tend to have diverging opinions about the attractiveness of most prospects; historically, company bids for individual offshore leases have varied widely. Consequently, the assumption that one company will be willing to invest in exploration is credible if the list of prospects was developed with attractiveness to industry in mind. However, this assumption would not apply to a group of prospects that represented an all-inclusive network of grid blocks over a broad area, because it is likely that some of these blocks would receive no bids. Consequently, leasing and exploration costs can be ignored in calculating economic levels of resources in a group of prospects and plays only if these costs were considered in the initial choice of prospects and plays to examine. A negative answer to the second question implies that the current tax structure treats exploration costs nearly the same way whether a project proceeds to development or is abandoned. If the tax system treats industry costs more generously upon abandonment than upon development, the potential developer will not ignore exploration costs in his “develop or abandon” decision, and neither should the assessor, even if exploration costs were considered in selecting the list of prospects and plays. If an assessor chooses to treat leasing and exploration costs as sunk, he should justify this decision by demonstrating that any tax differences between an abandoned prospect and a developed prospect are zero or small. RESOURCE ASSESSMENTS AND ENERGY POLICY Resource assessments become tools for policy analysts when issues of resource scarcity, environmental sensitivity of potential resource-bearing lands, or shortfalls in resource production need to be addressed. For example:

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures Government promotion of energy technologies may hinge on perceptions of the adequacy of the domestic fuels resource base. For example, perceptions of the size of the U.S. natural gas resource base will influence decisions about promoting combined cycle electricity generation fueled by natural gas. Proposals to stimulate drilling for oil and natural gas in the face of declining domestic production levels may hinge upon assessments of the conventional and unconventional or discovered and undiscovered part of the resource base. A conclusion that much larger volumes of natural gas reside in unconventional formations could shift policy attention to governmental assistance for production research, or to incentives aimed at tight gas or other unconventional resources. Similarly, assessment results that indicate large volumes of remaining recoverable oil in already discovered formations, with smaller volumes in the undiscovered portion of the resource base, might shift policy attention away from incentives for exploratory drilling and toward wider drilling incentives more likely to instill higher levels of infill and extension drilling. Policymakers evaluating proposals to sequester environmentally sensitive lands, like the Coastal Plain of the Arctic National Wildlife Refuge, will likely be strongly influenced by resource assessments covering the lands under review, the region (especially if an existing investment, like a pipeline, would lose value without continued development), and the nation. Policymakers are interested in resources primarily because they see them as a bridge to production. That is, they interpret an oil and gas resource assessment as a means of comprehending the prospects for maintaining high levels of domestic oil and gas production in the future. Because policymakers commonly are unsophisticated about resource terminology, they are likely to interpret a sharp change in resource volumes as signalling a similar change in future production prospects. Policymakers may find it difficult to comprehend—unless told very explicitly—that a so-called national assessment of oil and gas resources actually covers only a portion of the recoverable resource base: one that may have only a modest impact on future production. This is particularly true of the oil portion of assessments: fully 70 percent of the total U.S. oil reserve additions between 1979 and 1984 came from drilling thousands of extension and in fill wells in previously discovered oil fields, which are not part of the undiscovered resource base (Office of Technology Assessment, 1987). Although there is continuing debate among industry analysts about whether or not development drilling will continue to dominate reserve additions to this extent, there is no doubt that an assessment of only the undiscovered resource base cannot

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures adequately inform policymakers about prospects for future domestic oil and gas production from all sources. REFERENCES Brashear, J. P., A. Becker, K. Biglarbigi, and R. M. Ray. 1989. Incentives, technology, and EOR: potential for increasing oil recovery at lower oil prices. Journal of Petroleum Technology 41(2): 164-170. Energy, Mines and Resources Canada. 1977. Oil and Natural Gas Resources of Canada, 1976. Report 77-1. Ottawa, Canada. Kaufman, G. M. 1965. Statistical analysis of the size distribution of oil and gas fields. Pp. 109-131 in Symposium on Petroleum Economics and Evaluation. Richardson, Texas: Society of Petroleum Engineers. Kaufman, G. M., Y. Balcer, and D. Kruyt. 1975. A probabilistic model of oil and gas discovery. Pp. 113-142 in Studies in Geology No. 1: Methods of Estimating the Volume of Undiscovered Oil and Gas Resources. Tulsa, Oklahoma: American Association of Petroleum Geologists. Kuuskraa, V. A., F. Morra Jr., and M. L. Godec. 1987 Importance of cost-price relationships for least-cost oil and gas reserves. Society of Petroleum Engineers Paper No. 16289. Presentation to the Society of Petroleum Engineers Hydrocarbon Economics and Evaluation Symposium, March 1987, Dallas, Texas. Lee, P. J., and P. C. C. Wang. 1983a. Probabilistic formulation of a method for the evaluation of petroleum resources. Mathematical Geology 15: 163-181. Lee, P. J., and P. C. C. Wang. 1983b. Conditional analysis for petroleum resource evaluations. Mathematical Geology 15: 353-363. Lee, P. J. and P. C. C. Wang. 1985. Prediction of oil or gas pool sizes when the discovery record is available. Mathematical Geology 17: 95-113. Office of Technology Assessment. 1987. U.S. Oil Production: The Effect of Low Oil Prices— Special Report. OTA-E-348. Washington, D.C.: U.S. Congress. Oil and Gas Journal. 1989. U.S. fields with reserves exceeding 100 million barrels. January 30, pp. 69-70. Podruski, J. A., J. E. Barclay, A. P. Hamblin, P. J. Lee, K. G. Osadetz, R. M. Procter, and G. C. Taylor. 1988. Conventional Oil Resources of Western Canada. Paper 87-26. Ottawa, Canada. Roy, K.J. 1979. Hydrocarbon assessment using subjective probability and Monte Carlo methods. Pp. 279-290 in First IIASA Conference on Methods and Models for Assessing Energy Resources, M. Grenon, ed. New York: Pergamon Press. Roy, K. J., R. M. Procter, and R. G. McCrossan. 1975. Hydrocarbon assessment using subjective probability: probability methods in oil exploration. Pp. 56-60 in Probability Methods in Oil Exploration: American Association of Petroleum Geologists Research Symposium Notes, J. C. Davis, J. H. Doveton, and J. W. Harbaugh, eds. Tulsa, Oklahoma: American Association of Petroleum Geologists.

OCR for page 17
UNDISCOVERED OIL AND GAS RESOURCES:: An Evaluation of the Department of the Interior's 1989 Assessment Procedures U.S. Department of the Interior, Geological Survey and Minerals Management Service. 1989. Estimates of Undiscovered Conventional Oil and Gas Resources in the United States—A Part of the Nation's Energy Endowment. Washington, D.C.: U.S. Government Printing Office. U.S. Geological Survey. 1975. Geological Estimates of Undiscovered Recoverable Oil and Gas Resources in the United States. Circular 725. Washington, D.C.: Department of the Interior. U.S. Geological Survey. 1980. Future Supply of Oil and Gas from the Permian Basin of West Texas and Southeastern New Mexico. Circular 828. Washington, D.C.: U.S. Department of Interior. U.S. Geological Survey. 1981. Estimates of Undiscovered Recoverable Conventional Resources of Oil and Gas in the United States. Circular 860. Washington, D.C.: Department of the Interior. White, D. A. and H. M. Gehman. 1979. Methods of estimating oil and gas resources. American Association of Petroleum Geologists, Bulletin 63: 2183-2192.