SPE 63202 Adapting Probabilistic Methods to Conform to Regulatory Guidelines

Herman G. Acua, SPE, and D.R. Harrell, SPE, Ryder Scott Company Petroleum Consultants L.P.

Copyright 2000, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the 2000 SPE Annual Technical Conference and Exhibition held in Dallas, Texas, 14 October 2000. This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

address the increased interest in probabilistic analysis to estimate hydrocarbon reserves. Proved reserves were defined, in part, as those volumes of recoverable hydrocarbons with ... a high degree of confidence that the quantities will be recovered. If probabilistic methods are used, there should be at least a 90% probability that the quantities actually recovered will equal or exceed the estimate. This definition implies that satisfying the P-90 criteria is sufficient to define proved reserves. We will discuss later in this paper why defining proved reserves as the P-90 of any distribution is not always appropriate. Also, the definitions do not specify at what level the evaluator should apply the P-90 test (i.e. is it at the field level or the total portfolio level). Probable reserves were then described in the SPE/WPC definitions as those recoverable hydrocarbon volumes that ... are more likely than not to be recoverable. In this context, when probabilistic methods are used, there should be at least a 50% probability that the quantities actually recovered will equal or exceed the sum of estimated proved plus probable reserves. Possible reserves were defined as those recoverable hydrocarbon volumes that ... are less likely to be recoverable than probable reserves. In this context, when probabilistic methods are used, there should be at least a 10% probability that the quantities actually recovered will equal or exceed the sum of estimated proved plus probable plus possible reserves. The United States Securities and Exchange Commission (SEC) does not recognize probable and possible reserves. The SECs guidelines for reporting proved reserves are set forth in its Regulation S-X, Rule 4-10 and subsequent clarifying bulletins. In Regulation S-X, Rule 4-10 there are no guidelines for the interpretation of probabilistic analysis. The Regulation defines proved reserves as those recoverable hydrocarbon volumes with ... reasonable certainty to be recoverable in future years from known reservoirs ... Both the SPE/WPC and SEC proved reserve definitions have several other requirements that are usually applicable to

Abstract Probabilistic methods have introduced inconsistent interpretations of how to apply these methods while complying with reserve certification guidelines. The objective of this paper is to present and discuss some pitfalls commonly encountered in the application of probabilistic methods to evaluate reserves. Several Regulatory Guidelines that should be followed during the generation of recoverable hydrocarbon distributions are discussed. An example also is given to understand the evolution of reserve categories as a function of probabilities. Most of the conflicting reserve interpretations can be attributed to the current SPE/WPC reserve definitions where reserve categories are expressed in terms of probabilities of being achieved. For example, proved reserves are defined as those hydrocarbon volumes with at least a 90 percent probability of being equaled or exceeded (P-90). Unfortunately, these definitions alone fall short as guidance on how to derive the distributions from which these percentiles will be calculated. A simple example of this problem is the derivation of an exploratory prospect hydrocarbon resource distribution. While a P-90 can be calculated from this distribution, no proved reserves should be assigned to the prospect until it has actually been drilled and proven economical. Introduction In 1997 new reserve definitions were drafted and introduced by the SPE and the World Petroleum Congress (WPC). For the first time, these reserve definitions included some language to

H. G. ACUA, D. R. HARRELL

SPE 63202

deterministic methods that may conflict with probabilistic analysis if not properly incorporated. Evaluators of reserves should exercise caution when using probabilistic methods to ensure compliance with the definitions of reserves adopted by the SEC and the SPE/WPC. Caution is required because there are certain situations where indiscriminate application of probabilistic methods may produce results that are inconsistent with the reserve definitions. For example, the SEC definition of proved reserves does not explicitly recognize the use of the probabilistic method, and in no way allows for the probabilistic method to be used in such a manner as to violate any term of that definition. In this paper, we will first present a short definition of probabilistic analysis and the risks and benefits of using this technique. Next, we will address some significant shortcomings in the current reserve definitions, and then, we will present some examples on how some of these shortcomings can be addressed in the evaluation of reserves. Discussion of Probabilistic Analysis of Reserves The probabilistic analysis of reserves relies on the use of probabilistic techniques to estimate the uncertainty of the recoverable hydrocarbon volumes. In its purest sense, these probabilistic methods are used to collect and organize, evaluate, present and summarize data. These methods provide the tools to analyze large amounts of representative data so that the significance of that datas variability and dependability can be measured and understood. Probabilistic analysis should be considered an important tool for internal company analysis to understand and rank its hydrocarbon reserves and resources and associated risks. This method provides the tools to identify the upside and the downside hydrocarbon potential to better organize the companys portfolio and to more efficiently allocate capital and manpower resources. However, it should be understood that the objectives of a hydrocarbon property ranking study and a SPE/WPC or SEC reserve reporting evaluation might be different. For example, companies may have their own guidelines to group and analyze hydrocarbon assets to allocate company resources or for property acquisitions. These company guidelines may vary from project to project or year to year (depending on pricing assumptions) and may be different from those guidelines provided in the SPE/WPC and SEC definitions. It becomes then the primary challenge of the evaluator to reconcile both evaluations. E.C. Capen1 discussed this issue of having different objectives and different audiences for the evaluation of reserves. He correctly remarked that the evaluation needs of the explorationists, geologists and geophysicists may not
1

necessarily be the same as the needs arising from lending institutions or stockholders. While the reserve evaluation should allow the explorationists ... to dream a little but still hold them accountable, the lenders should be able to ... sleep well at night knowing their money takes no inordinate risks. This is an obvious conclusion since the lenders exposure to the downside risk is much greater than their exposure to the upside. Lending institutions do not particularly benefit from having large project upsides with low probabilities since they generally lend money at fixed rates. It, thus, follows that lending institutions are much more interested in the quantification of the downside risks and not so for the upside. About the stockholder, Capen wrote that he ... cannot afford biased estimates lest the buy and sell signals get all confused and cost him money. Probabilistic analysis is not a substitute for the reserve guidelines adopted by the SPE/WPC and the SEC. A common but, in our opinion, unacceptable trend in reserve evaluations has been to rely on probabilistic evaluations to boost reserve values if they cannot be supported by deterministic methods. This generally results from the aggregation of various distributions as will be discussed later in this paper. Probabilistic methods should not be considered as a sequence of simple mechanical steps to reach the desired outcomes. In fact, probabilistic analysis requires as much, if not more, good judgement than deterministic methods for assigning reserves. Probabilistic methods are most suitable for early hydrocarbon resource/reserve evaluation. They provide flexible modeling techniques to evaluate variable interactions and outcome options. Therefore, these methods require a customized approach to correctly model the variable interactions. Unfortunately, the great flexibility of probabilistic analysis makes this method easy to abuse. Restricting the probabilistic analysis to the guidelines provided by the reserve definitions can help to standardize the results for reserve reporting purposes. In the next section we first explore some of the shortcomings of the current definitions and then we will present some examples for the practical application of reserve definitions and probabilistic analysis. Conforming with the Reserve Definitions The primary factor for inconsistent reserve reporting evaluations is the description of reserves as percentiles of a distribution. In the current SPE/WPC definitions, proved reserves are defined as P-90, the proved plus probable as P-50, and the proved plus probable plus possible (3P) as P-10. These percentiles for the different reserve categories have been interpreted by some as finally quantifying the meaning of the phrases, high degree of confidence, more likely than not and less likely to be recoverable than probable reserves. The industry was quick to adopt these percentile definitions at face value. Unfortunately, the SPE/WPC

E. C. Capen: A Consistent Probabilistic Definition of Reserves, SPE Reservoir Engineering Journal, February 1996, pp 23-28.

SPE 63202

ADAPTING PROBABILISTIC METHODS TO CONFORM TO REGULATORY GUIDELINES

definitions do not provide direct guidance on how the distributions are to be generated. Consequently, what may be a P-90 to one evaluator may be something else to another evaluator. The definitions state without qualification that the percentile tests are sufficient for the establishment of reserves if the probabilistic method is used, irrespective of other aspects of the definitions. Thus, the definitions permit the calculation of at least two values of reserves, those estimated by the deterministic method (honoring all terms of the definitions) and those estimated by the probabilistic method. Figure 1 illustrates this point. Figure 1: Example of Early Reserve Assessment

From Figure 2, one can see that there are different values of recoverable oil that fit the definition of P-90 for proved reserves. While it is important to generate and understand the potential for the entire structure, it is obvious from this simple example that for reserve reporting purposes the definition of proved reserves based on a P-90 definition may not be enough. For reserve reporting, one also should incorporate all the guidelines that traditionally may be associated with deterministic methods. For example, can the structure on the opposite side of the tested fault block be considered as a known reservoir? If not, should it be left out of the analysis of proved reserves? Similarly, the volumes of rock below the LKO may not be included. Application of these guidelines would then result in the determination of proved reserves from the first distribution in Figure 2. This distribution is what we would call the proved reserve distribution. Subsequently, the question is whether after restricting the distribution to account for various reserve guidelines, one should still use a P-90 to estimate the proved reserves. To answer this, let us first examine the approach that would have been taken in a deterministic analysis. In a deterministic analysis, the evaluator would have restricted the reservoir area to conform with the LKO guideline. Then weighted average values of porosity, water saturation, net-togross ratios and recovery factors would have been applied to estimate the proved reserves. The deterministic analysis is therefore arriving at the expected value of reserves of the proved reserve distribution. Therefore, in probabilistic analysis, the proved reserves may be represented better by the P-50 of the proved reserve distribution than the P-90 of that distribution. However, notice in Figure 2, that the P-50 of the proved reserve distribution may still only represent a small fraction, P-98 or less, of the entire recoverable oil distribution. Two important conclusions can be made from this example. First, special considerations need to be made when constructing the distribution of recoverable hydrocarbons to ensure that the definition guidelines are being honored. Second, the actual percentile assigned to each of the reserve categories may be different depending on which distribution they apply to. Moreover, the distributions of recoverable hydrocarbons need not be the same for each reserve category. There are currently at least two acceptable alternatives in the evaluation of proved reserves using probabilistic analysis to conform with the regulatory guidelines. The first alternative would be to proceed as shown in the prior example. That is to first derive a proved reserve distribution by conforming to certain variables that cannot be subjected to speculation based on regulations. Then, selecting the P-50 value from this distribution to estimate proved reserves. After estimating the proved reserves in this way, the percentile corresponding to proved reserves from the overall reserve and

For discussion purposes let us assume that there is 3-D seismic data over the entire structure shown in Figure 1 and that one well has been recently drilled and tested oil at economical rates. The location of the well is represented by the dot in Figure 1. The lowest known oil (LKO) is represented by the first contour around the well. The total area on both sides of the fault represents the area that may be productive as indicated by seismic analysis. The different contours represent various degrees of confidence that the reservoir may be productive from the seismic and geologic analysis. A distribution of recoverable oil can then be generated to estimate the reserves attributable to this structure. Figure 2 shows such a distribution. Figure 2: Example of Recoverable Hydrocarbons Distribution Cumulative Probability P-90

P-50

P-10 Recoverable Oil

H. G. ACUA, D. R. HARRELL

SPE 63202

resource base can be backcalculated from a second distribution that has not been constrained by the proved guidelines. If the opportunity for upside is large or if the field is in the very early stages of development, reportable proved reserves may only represent a small fraction of the overall distribution (i.e., a percentile higher than P-90). On the other hand, if the current stage of field development is such that the proved distribution is the same as the overall distribution, then proved reserves may be better represented by the P-50 of that distribution. In fact, it follows from this discussion that the percentiles applicable to each of the reserve categories are dynamic throughout the life of the reservoir and are not a fixed value. Obviously, during the early field development stages reportable proved reserves may represent only a small percentile of the overall reserve and resource opportunity. With increased development and decreased uncertainty, proved reserves may approach a P-50 or expected value. The second approach is similar to what has been discussed, except that proved reserves are estimated deterministically.23 That value would then be compared to the unrestricted distribution of recoverable hydrocarbons to determine its corresponding percentile. By determining the proved reserves deterministically, conforming with the regulation guidelines and using expected values of rock properties and engineering factors, the evaluator is basically determining the expected value of the proved distribution as previously defined. Evaluation Variables In general when estimating reserves, the evaluator deals with three types of variables: 1) Geologic, 2) Engineering and 3) Economic. Each of these variables may have degrees of uncertainty that may be represented as distributions. This paper does not deal with the aspects of generating the random variable distributions or the correlations between them. Nor does this paper address the issues of properly incorporating the reservoir physics to represent the actual producing mechanisms in the probabilistic model. Obviously, the evaluator should be familiar with how the data were collected and what they represent. Also, the reader should be aware that throughout the discussion of reservoir variables, scale consistency between the variable distributions and the model to which they apply should be maintained. This means,
2

for example, that if the exercise at hand is to evaluate field level reserve potential, then the distributions of reservoir properties should be the distribution of average field properties. This paper deals, however, with the restrictions that may be imposed in a probabilistic evaluation by the definitions. As previously stated, certain variables may not be represented in a distribution fashion for reserve reporting purposes. The objective of the definitions is not to limit the understanding of the hydrocarbon assets, but rather to ensure that the reported reserves, in particular proved reserves, are evaluated consistently. For example, some of the geologic variables to consider may be: Water-Hydrocarbon Contact Gas-Oil Contact Bulk Rock Volume Net-to-gross ratios Porosity Water Saturation Cutoff values Reservoir Facies Distribution The first geologic variable listed in bold letters above is directly affected by the definition of reserves. While there may be some uncertainty as to the exact location of the fluid contacts, the reserve definitions are very clear in that for proved reserves fluid contacts should not be speculated upon. Therefore, under the current definitions, the generation of the proved hydrocarbon-in-place distribution should incorporate the LKO value and not a distribution of the contacts. However, it may be reasonable to expect that more hydrocarbons may exist below the LKO depth. Engineering examples of variables that may be directly affected by the reserve definitions are: 1) well spacing, 2) recovery factors, and 3) implementation of secondary and tertiary projects. While downspacing or the implementation of secondary projects may result in increased recoveries, the definitions prevent us from speculating on these results unless there have been successful tests or pilots. Therefore, it would be an inconsistent application of these guidelines if the incorporation of these operating practices in the probabilistic analysis increases the estimate of proved reserves. Reserve distributions that incorporate the results from waterflooding projects yet to be tested through pilots may result in overstating proved reserves, as currently defined, particularly during the aggregation of multiple field reserves. The subject of reserve aggregations is discussed in the next section. The reserve definitions are even less flexible when it comes to economic parameters. To categorize reserves as proved, those reserves have to be economical under current economic conditions. While it may be appropriate to incorporate

Patricelli, J.A., McMichael, C.L.: An Integrated Deterministic/Probabilistic Approach to Reserve Estimations, JPT (Jan. 1995) pp. 49-53. 3 Nangea, A.G., Hunt E. J.: An Integrated Deterministic/Probabilistic Approach to Reserve Estimation: An Update, SPE 38803, presented at the 1997 SPE Annual Technical Conference and Exhibition held in San Antonio, Texas, 5-8 October 1997.

SPE 63202

ADAPTING PROBABILISTIC METHODS TO CONFORM TO REGULATORY GUIDELINES

uncertainty and volatility in hydrocarbon prices, operating costs and development costs in probabilistic analysis, these variables are generally fixed in the reserve definitions. A typical example of a proved reserves economical evaluation that may not be appropriate under the current reserve definitions is the consideration of a learning curve. We are all familiar with the concept that, in general, as we do tasks repeatedly we get better at it. For example, a drilling learning curve, which shows a reduction in drilling costs over time. A probabilistic analysis can be made from past company performance to define a function of drilling cost savings, in percentages, with increased development. While this type of analysis is important for internal company reporting and analysis, it is not appropriate for reserve reporting unless the cost savings trend has been clearly established in the subject field. Otherwise, the appropriate development costs to use for proved reserve reporting are the actual costs regardless of expectations of future savings. While dealing with projects outside of North America, the evaluator generally has to make assumptions regarding the timing for infrastructure, such as pipelines and market conditions, particularly for gas. While these considerations are important for company planning purposes and can be incorporated in risk economic models, they may prevent the classification of proved reserves. In general, there must be established infrastructure and product markets to classify reserves as proved for SEC purposes. To illustrate this problem, consider for example five oil fields on primary depletion with recovery factors ranging between 12 and 18 percent for each field. As shown in Figure 3, these recovery factors can be expressed as triangular distributions with a P-90 of 12 percent and a P-10 equal to 18 percent with a mean value of 15 percent. Figure 3: Primary Recovery Factors

Figure 4: Secondary Recovery Factors

0.00

0.04

Recovery Factor

0.08

0.12

0.15

The primary plus secondary recovery factors can, thus, be obtained for each probabilistic iteration as the summation of the two distributions. Figure 5 shows the aggregated recovery factors for all the fields. Figure 5: Primary Plus Secondary Recovery Factors
Probability
.026

.019

.013

.006

.000 0.18 0.21 0.23 0.26 0.28

Recovery Factor

Multiplying the original oil in place volumes by the primary plus secondary recovery factors could then calculate reserve estimates. The aggregated percentile recovery factors for all five fields are shown in Table 1. Table 1: Portfolio Primary plus Secondary Recovery Factors Recovery Factor (%) 20.5 21.3 21.8 22.3 22.7 23.2 23.7 24.2 24.9

P-90 P-80 P-70 P-60 P-50 P-40 P-30 P-20 P-10

0.10

0.12

0.15

0.18

0.20

Recovery Factor

These fields are considered good candidates for secondary recovery but no pilot has been implemented yet. The additional recovery factors due to secondary projects are estimated between 0 to 12 percent. These secondary recovery factors can also be represented by triangular distributions with a minimum of 0 and a P-10 of 12 percent as shown in Figure 4.

Notice that the P-90 from the aggregated recovery factor distribution is higher than the P-10 of the primary alone. Obviously, disclosing proved aggregated reserves with an equivalent recovery factor of 20.5 percent would not be appropriate in this case. Although the aggregated recovery factor distribution shown in Figure 5 and Table 1 may be the correct outcome if all five fields are waterflooded, these expectations can not be used to book proved reserves at the current field development stages and under the current reserve definitions. So what is the

H. G. ACUA, D. R. HARRELL

SPE 63202

solution to this problem? Obviously, proved reserves should be based only on reserve distributions that have been derived exclusively from primary recovery factors until the appropriate waterflooding pilots have been implemented and response has been verified. One may argue that calculating proved reserves based on primary recovery factors in this example underestimates the overall potential of the fields. Someone else may point out that this conservatism is the reason for the reserve growth observed with added development. While these arguments may be correct, one should keep in mind that the reserve categories have meaning only within the context of their definitions. Therefore in this example, proved reserves have meaning only for those volumes calculated without consideration of secondary recovery. Under the current definitions, secondary reserves without a successful pilot should be excluded from the proved category regardless of the waterflooding perspectivities of the field. Of course, the reservoir engineer should make those calculations to allocate company resources and funds. This brings us back to our original statement that during the early stages of field development proved reserves may be represented by percentile values of the overall reserve distribution higher than the P-90. If we return to the original postulation of this paper to first determine individual distributions that comply with the reserve definitions and then select the P-50 from those distributions to estimate the proved reserves, then the summation of those P50 values should be very similar to the P-50 of their aggregate. Additionally, reserve distributions derived using the definitions for proved should be risk-equivalent and, thus, compatible for aggregation without the problems discussed in the next section. After determining the proved reserves, one can determine the appropriate percentile of proved reserves within the overall company asset reserve distribution, and maybe even disclose or discuss this percentile and its relationship to company upside. Probable and possible reserves may be estimated from P-50 and P-10 of the overall aggregated distribution of reserves (excludes exploratory assets). In conclusion, the application of probabilistic analysis and the variable distributions used thereof should not conflict with the guidelines established by the definitions. These definitions were established to standardize, as much as possible, reserve reporting. This is not to say that the current reserve definitions should not evolve. On the contrary, these definitions should and will continue to evolve to address these issues and other issues. Capen1 wrote As long as estimators provide the entire distribution, others can define reserves however they choose as long as they pick some point in the distribution. While this may be applicable for internal company reserve and budgeting purposes, this statement is obviously unacceptable

for the purpose of reserve reporting. No lending institution, market analyst or shareholder would be able to conduct meaningful comparative reserves and income evaluations between E&P companies without the standards. Dry Hole Risk Factor and Aggregation of Distributions One aspect that is often overlooked during the aggregation of distributions is the fact that distributions may have different risk levels. Consistent with the accepted practice that the summation of proved, probable and possible deterministic reserves is not appropriate without risk adjusting each reserve category first, reserve distributions should not be aggregated unless they have been adjusted for their dry hole factor. Consider for example the three fault blocks in Figure 6. Figure 6: Aggregation of Distributions

I A

LKG

B
Only fault block A has been drilled at this stage of field development and the well proved gas to a lowest known depth (LKG). Seismic data analysis indicates the possibility of additional gas down to the deepest shaded area. It may be desirable then to aggregate the estimated recoverable gas volumes from all three fault blocks. Obviously, fault blocks B and I have a higher risk than fault block A. Distributions for bulk rock volumes, net-to-gross ratios, porosity, gas saturation, and recovery factors are used to generate recoverable gas volumes for each fault block. Each property distribution is based on fault block averages expected in this area and for the specific reservoir age.

SPE 63202

ADAPTING PROBABILISTIC METHODS TO CONFORM TO REGULATORY GUIDELINES

The individual reservoir variable distributions are applied to the volumetric gas equation to estimate the original gas in place. The original gas-in-place is then multiplied by a recovery factor, sampled from the recovery factor distribution, to obtain the recoverable gas for each fault block. These values are shown in Table 2. These recoverable gas volumes reflect only the uncertainty in the reservoir properties and performance and, therefore, are representative of the success cases only. Although the information obtained this way may provide an important perspective to the overall upside of the prospect, there is no reason why the probability of dry hole shouldnt be incorporated into the probabilistic analysis and the aggregated distribution of recoverable gas volumes. Table 2: Recoverable Gas Volumes (MMscf) P-90 P-80 P-70 P-60 P-50 P-40 P-30 P-20 P-10 Fault Block A 9,658 11,517 12,818 13,943 14,971 16,002 17,102 18,352 20,146 Fault Block B 13,124 15,655 17,539 19,083 20,493 21,909 23,588 25,265 27,500 Fault Block I 4,904 5,852 6,508 7,084 7,629 8,188 8,744 9,480 10,475

Figure 8: Comparison of Aggregated Distributions Cumulative Probability

Recoverable Gas (MMscf) Table 3 compares the unrisked and risked aggregated gas recoverable volumes from these three fault blocks. Table 3: Comparison of Recoverable Gas Volumes (MMscf) P-90 P-80 P-70 P-60 P-50 P-40 P-30 P-20 P-10 Unrisked 33,731 36,909 39,301 41,184 42,923 44,954 46,932 49,299 52,286 Risked 16,864 22,963 28,141 32,498 35,658 38,553 41,279 44,357 48,534

The aggregation of these three distributions results in the distribution of unrisked recoverable gas shown in Figure 7, which assumes that all three fault blocks are successful. Figure 7: Unrisked Aggregation of Fault Blocks A, B and I

Cumulative Probability
1.000 .750

While the prior discussion of risk and aggregation of distributions may be obvious to some, it is still common practice to see reserve distributions aggregated without consideration for risk. The fact that data are presented in distribution form can make it counterintuitive that additional risking may be necessary prior to aggregation. The amount of recoverable gas that can be booked as proved, probable and possible for this example will be discussed later in this paper. Aggregation of Reserve Portfolio

.500

.250 Mean = 43,091 .000 20,000 31,250 42,500 53,750 65,000

Recoverable Gas (MMscf) To calculate the risked and, therefore, more representative recoverable gas distribution at this stage of development, one should apply the individual chances of success (COS) for each fault block during the simulation. Figure 8 compares the unrisked distribution with the risked distribution obtained by assuming a COS of 100 percent in fault block A, 75 percent in fault block B and 50 percent in Fault block I.

The aggregation of company reserves has traditionally been fairly straightforward. The summation of all individual proved reserves within its portfolio provide the total company proved reserves. Similar arithmetic is performed for probable and possible reserves. This process has been slightly complicated by the introduction of probabilistic evaluation of reserves and the current reserve definitions. It has been well established that the summation of individual percentiles from various distributions is not equal to the same percentile of the aggregate. Thus it follows that adding up the P-90, as defined for proved reserves, from the various reserve

H. G. ACUA, D. R. HARRELL

SPE 63202

portfolio components will not be equivalent to the P-90 of the aggregated total. In fact, the summation of the P-90 will be, in general depending on the number of uncorrelatable opportunities in the portfolio, much lower than the P-90 of the aggregate. On the other hand, the summation of the P-10 will tend to exaggerate the upside and be much higher than the P10 from the aggregated distribution. The reason for this is very simple -- the larger the amount of opportunities available to a company, the more it can diversify risk. In other words, independent opportunities will tend to compensate each other; some may be low and some may be high, thus reducing the uncertainty of the total portfolio outcome. The addition of individual P-50 values should approach the P-50 of the aggregated distribution, as long as the median and the mean are not too dissimilar. The obvious dilemma becomes how to report individual and total company portfolio reserves. To make matters worse, each distribution in the reserve portfolio needs to be properly risked and may or may not comply with all the definitions. The current definitions provide no wording for the proper method for aggregating reserves; neither do the definitions provide guidance for the appropriate level of aggregation to apply the percentiles associated with each reserve category. Consider, for illustration purposes, a company with only probable reserves from ten fields defined at the field level at the P-50s. Recoverable hydrocarbon volumes above P-55 are zero. That is, the prospects may have a COS of 55 percent. The following table shows the recoverable oil volumes for one field and the aggregates of two, three, five and all ten fields. Table 4: Field Reserve Aggregation Example (MMBbl) 1 Field 0.0 0.0 0.0 0.0 10.9 13.2 14.7 16.2 18.0 2 Fields 0.0 0.0 12.4 14.3 15.9 17.8 21.4 28.2 31.9 3 Fields 9.6 13.5 15.9 19.2 26.1 29.4 32.3 35.9 43.6 5 Fields 16.6 27.1 31.3 36.2 42.0 45.8 50.4 56.7 64.2 10 Fields 50.5 61.3 69.5 76.4 82.7 88.8 94.9 102.8 114.3

explain or disclose this risk to their investors? All these are real and difficult questions without simple solutions. As we have mentioned earlier, the definitions provide no guidance for the process of aggregating portfolios. For reserve aggregation, we emphasize one more time that the evaluator should consider the audience for the report. Clearly, this type of analysis is of great value for the companys internal resource allocation and portfolio risk analysis. It helps the company understand the diversification of their portfolio. It provides an insight to the impact that divesting fields and properties may have on their overall risk profile. It helps the company measure the level of commitment necessary to reach a certain level of success. However, for reserve reporting purposes and under the current definitions it may be difficult if not impossible to justify proved reserves in this example. Under SEC guidelines, we should not disclose any proved reserves. The P-90 volumes of the aggregated totals can not be allocated with reasonable certainty to any specific field or prospect. In fact, to realize the P-90 volumes (or any other percentile volume of the aggregated totals), the company would have to be committed to the development of all ten fields regardless of early disappointments or successes. The SEC/WPC provides no guidance to handle this type of problem probabilistically. However, the only justification to book proved reserves under the SPE/WPC guidelines would be the fact that the aggregated P-90 value has a positive value. This, however, would conflict with the grain of the remaining guidelines. Disclosing the aggregated volumes in this case as proved may leave investors unprotected against portfolio changes, divestitures, delayed or incomplete portfolio development. On the other hand, not presenting this information at all does not adequately reflect the upside of a companys portfolio. Until the reserve definitions are modified to accommodate this type of problem and the investing public has been familiarized on this subject, we would recommend against disclosing aggregated proved reserves that can not be readily identified at the individual field level. However, we would recommend discussing these volumes and the overall risk profile of the company that results from its existing portfolio in company filings and reports. This allows investors to make their own decisions based on their comfort level about the overall companys upside without the assumption of reasonable certainty associated with the proved reserve category. Example Problem

P-90 P-80 P-70 P-60 P-50 P-40 P-30 P-20 P-10

This simple example highlights the problem that one encounters when aggregating multiple fields. Although at the individual field level none of the fields have proved reserves, the aggregated total shows significant reserves at the P-90 level for three or more fields. Should this company then be in a position of disclosing the aggregated reserves as proved? If so, how should this information be disclosed to the investors and shareholders? What if the company decides to divest individual fields or all of them? Prospective buyers may not pay anything for proved reserves for the individual fields. How can the company then

In this example, we will walk through the analysis of reserves as a gas exploratory prospect is first drilled. Figure 9 shows the Gross Hydrocarbon Isochor for this example.

SPE 63202

ADAPTING PROBABILISTIC METHODS TO CONFORM TO REGULATORY GUIDELINES

Figure 9: Gross Hydrocarbon Isochor

distribution carries a much higher risk than a P-90 from a gas reserves distribution. The salt dome prospect is not yet considered a known reservoir and Table 5 shows the result for a successful case without dryhole consideration. A similar table, after applying the risk factors to each fault block through correlated binomial distribution of the COS, could be generated to get the risked recoverable gas distribution. However, this still would not be enough to elevate any of the gas volumes to a reserve category. The company then successfully drilled fault block A to test the exploration concept. The well A-1 found gas down to the base of the sand thus establishing the LKG for that fault block. Seismic analysis indicates the possibility of more gas downdip of the LKG, but the seismic analysis is not conclusive. This is shown schematically in Figure 6. Knowing that we have 9 prospective fault blocks with one already tested, the issue becomes now how much proved, probable and possible reserves can be booked.

The exploration prospect consists of 9 fault blocks labeled A through I. The first step involves the determination of the distributions for the gross hydrocarbon bulk rock volume, netto-gross ratios, porosities, water saturations, formation volume factors and recovery factors. These distributions and their appropriate correlations are used in the volumetric equation to derive original and recoverable gas distributions for each fault block. This is shown schematically in Figure 10. This figure also shows the type of distribution used for each variable. Figure 10: Gas Volumetric Equation
BRV
BR V A

NTG
NT G A

POR
P o ro sity A

Sw
SW A

Following our prior discussion, the first step would be to establish a distribution of recoverable gas that complies with the definitions for fault block A. We will call this distribution the proved distribution. In this case, the proved distribution was restricted by the LKG. There may still be some uncertainty associated with the petrophysical and structural elements of fault block A that would make the generation of the proved distribution necessary. Otherwise, a deterministic evaluation of that fault block may be enough. The second step is to generate the distribution for the entire fault block A recoverable gas. This distribution would include the upside that is indicated by the seismic below the LKG. Table 6 compares the recoverable volumes of gas from the two distributions. The second column shows the recoverable volumes of gas estimated from the proved distribution for each of the percentiles in column one. The third column shows the total distribution of recoverable gas. Table 6: Comparison of Recoverable Gas for Fault Block A Proved to LKG (MMscf) P-90 P-80 P-70 P-60 P-50 P-40 P-30 P-20 P-10 3,920 4,628 5,184 5,619 6,007 6,411 6,842 7,286 7,982 Entire Fault Block A After Drilling (MMscf) 9,658 11,517 12,818 13,943 14,971 16,002 17,102 18,352 20,146

43,560 OGIP =

*
7 2,6 2 2 8 0,3 4 2 8 8,0 6 1 9 5,7 8 1 1 03 ,5 00

*
0 .6 5 0 0 .7 4 0 0 .8 3 0 0 .9 2 0 1 .0 1 0

*
0 .1 1 0 0 .1 7 0 0 .2 3 0 0 .2 9 0 0 .3 5 0

* (10 .0 9 0 .2 9 0 .4 9 0 .6 9 0 .8 9

BG A

0 .0 2 8

0 .0 3 0

0 .0 3 1

0 .0 3 3

0 .0 3 4

Bg The results from this exercise then provide the distribution of gas resources by aggregating the recoverable volumes from all the fault blocks. Table 5: Salt Dome Example Unrisked Recoverable Gas MMscf 118,288 123,751 127,762 131,104 134,323 137,655 141,116 145,033 150,801

P-90 P-80 P-70 P-60 P-50 P-40 P-30 P-20 P-10

From the stage of field development it is obvious that there are no bookable reserves for this prospect. Although we have derived a P-90 for a gas resources distribution, the P-90 of this

Several observations can be made from Table 6 while determining the appropriate amount of reserves to be booked from this fault block. First, notice that the P-10 from the

10

H. G. ACUA, D. R. HARRELL

SPE 63202

proved distribution is less than the P-90 of the entire fault block as defined from seismic. This results from the well location, which is not the subject of the discussion. But if we were to use the P-90 from the second distribution to determine the proved reserves, we would obviously be stretching the volumes beyond those proved by the bit. Since this fault block will most likely be a one well fault block, only performance will be able to prove or disprove the geologic and seismic assumptions that define the upside. Given the fact that there is no production data to verify the upside from this fault block, one then has to look at the proved reserve distribution. Consistent with our prior discussion, proved reserves would be defined as the P-50 recoverable gas volumes from this distribution. In this case, the 6,007 MMscf are the expected recoverable gas volumes from the proved distribution and are, thus, the proved reserves. The proved plus probable reserves would be defined as the P-50 of the entire fault block A distribution and the 3P reserves (proved plus probable plus possible) would then be estimated from the P-10 of the same distribution. Table 7 then shows the reserves per category for this fault block. Table 7: Gas Reserves Fault Block A Fault Block A (MMscf) 6,007 8,964 5,175

As we discussed earlier, these distributions are not riskconsistent since fault block A has a penetration while the other two do not. During the aggregation of these three distributions one should apply the COS to determine the total probable and possible reserves from all three blocks. This process was discussed earlier in the section dealing with risk and aggregation of distributions. Assuming a COS of 100 percent for fault block A, 75 percent for fault block B and 50 percent for fault block I, the aggregated recoverable gas distribution shown in Table 9 is obtained. From this table the three fault block probable reserves are determined from the P50. Possible reserves for the same blocks are estimated from the P-10 of the same risk adjusted distribution. Table 9: Risked Recoverable Gas Fault Blocks A, B and I MMscf 16,864 22,963 28,141 32,498 35,658 38,553 41,279 44,357 48,534

Proved Probable Possible

P-90 P-80 P-70 P-60 P-50 P-40 P-30 P-20 P-10

After determining the appropriate reserve levels for fault block A, the question remains whether additional reserves can be booked for the remainder of the structure. Statistical information on fault block success ratio in tested structures may be of some use if available. Lets first examine the two fault blocks adjacent to fault block A. These blocks are fault blocks B and I in Figure 6. In our opinion, it would not be appropriate to book proved reserves for these fault blocks at this stage since the structures have not yet been penetrated by a well. Probable and possible reserves can be booked based on seismic analysis and analogy to fault block A. Table 8 shows the unrisked recoverable gas for fault block A and each of the adjacent fault blocks. Table 8: Recoverable Gas Volumes (MMscf) Fault Block A 9,658 11,517 12,818 13,943 14,971 16,002 17,102 18,352 20,146 Fault Block B 13,124 15,655 17,539 19,083 20,493 21,909 23,588 25,265 27,500 Fault Block I 4,904 5,852 6,508 7,084 7,629 8,188 8,744 9,480 10,475

Whether or not reserves can be assigned to any of the remaining 5 blocks will depend on the certainty of the seismic analysis and the fault block COS observed in this gas province. Despite the mechanics of probabilistic evaluations, the experience and good judgement from the evaluator are still necessary and valuable. In this example, no reserves were assigned beyond the two adjacent fault blocks resulting in total reserves shown in Table 10. Table 10: Total Gas Reserves Fault Block A (MMscf) 6,007 29,651 12,876

Proved Probable Possible

P-90 P-80 P-70 P-60 P-50 P-40 P-30 P-20 P-10

As a final analysis, one can compare the current reserve base with the overall unrisked and risked gas resource distribution to better understand the additional potential and prioritize company resources. The unrisked distribution shows the values of recoverable gas if all 9 fault blocks are successful. The risked distribution was generated with a relatively high (75 percent) correlation coefficient between adjacent fault blocks COS. Table 11 shows the total recoverable gas volumes for the entire structure.

SPE 63202

ADAPTING PROBABILISTIC METHODS TO CONFORM TO REGULATORY GUIDELINES

11

Table 11: Total Recoverable Gas Volumes (MMscf) 7. P-90 P-80 P-70 P-60 P-50 P-40 P-30 P-20 P-10 Unrisked 118,018 123,711 127,621 131,193 134,681 137,929 141,361 145,432 150,695 Risked 33,241 44,390 56,310 67,592 79,533 90,543 102,817 113,558 127,563 The percentiles applicable to proved, probable and possible reserves are dynamic throughout the life of the reservoir. Recoverable hydrocarbon volume distributions may not be risk consistent. These distributions need to be adjusted for risk (COS) prior to aggregation. One approach to define proved reserves is to select the P50 value from a proved distribution. Where a proved distribution is defined as the distribution of recoverable hydrocarbons derived by strictly following the definitions and regulatory requirements.

8.

9.

Notice that the unrisked 3P value of reserves of 48,534 MMscf corresponds to approximately a P-75 of the total risked recoverable gas distribution. This information is very indicative of the significant upside, beyond current bookable reserves, that can be achieved with further development. Conclusions In this paper, we have discussed how the reserve definitions may affect the implementation of probabilistic methods. The following conclusions can be made from the preceding discussions: 1. Probabilistic methods should not be used as a tool to boost reserve reporting. Probabilistic methods are most useful for the prioritization of the companys portfolio and the allocation of capital and manpower resources. The evaluation of hydrocarbon resources and reserves may have different objectives and parameters for different audiences. The SEC set those parameters for proved reserve-reporting purposes through their regulations and guidelines. One should keep in mind that the reserve categories have meaning only within the context of their definitions. Therefore, even when the evaluator expects upside recoverable hydrocarbon volumes, those volumes can not be reported if they fall outside the definitions. This ensures consistent reserve reporting among companies. SEC regulations and SPE/WPC reserve definitions may prevent certain variables from being expressed in distribution form during the probabilistic analysis. Satisfying the P-90, P-50 and P-10 from a hydrocarbon resource distribution is not enough to book reserves. The SPE/WPC reserve definitions do not provide guidelines to the appropriate level to which the P-90, P-50 and P-10 are applicable. Also, these definitions do not provide guidelines for the aggregation of prospects. Therefore, extreme caution should be exercised when reporting aggregated volumes as reserves.

2.

3.

4.

5.

6.