Author’s note: this article has been heavily redacted since its original publish date. Content on the upstream business was redacted and re-posted elsewhere in order to expand more on the business fundamentals there and the resource fundamentals here.
- The previous installment established that cost-of-supply is the overwhelming driver of petroleum exploration and production value.
- The geological processes which resulted in the accumulation of hydrocarbons and resulted in the formation of petroleum reservoirs strongly influence the quantities of recoverable resources and their production characteristics.
- Additionally, geology is a key determinant of cost, and therefore also a key driver of upstream value.
- A grasp of geological concepts facilitates the interpretation of language within company disclosures — ultimately helping investors identify instances where value and price diverge.
Figure 1: A Different Kind of Lease
Source: Art and Framing Plus
Part 1 of this series broadly addressed the fundamentals of the broader petroleum value chain, especially from an investor’s perspective. This installment deep dives on the fundamentals of economic geology (i.e., petroleum resource management) in order to impart a holistic view of geological and technical factors governing petroleum recovery. Since cost-of-supply is the overwhelming driver of value in the upstream oil and gas business, and geology is often the overwhelming factor underlying cost, a basic understanding of petroleum geology is necessary to fully grasp the economic drivers. Topics include petroleum geology, petroleum geography, resource classification, petroleum recovery, and the fundamentals of resource quantity and production estimation. Following installments will leverage this knowledge to address the business fundamental of exploration and production, and subsequently the economics of the upstream business. At a later point, these foundations in petroleum geology, business fundamentals, and economics will help us maximize the utility of financial reports and unravel accounting minutiae.
Petroleum is a special type of hydrocarbon with two critical functions in society: as a primary fuel source; and as the hydrocarbon building block of choice used in a myriad of applications (e.g., plastics, waxes, paints, textiles, cosmetics, etcetera).
Petroleum resources are accumulations of hydrocarbons which exist naturally within the Earth’s crust in both liquid and gaseous forms. Petroleum liquids include various grades of crude oil (which can vary significantly in density, sulphur content, and acidity), condensate (i.e., very light oil), natural gas liquids (NGLs; reservoir gases which exists in liquid form on the Earth’s surface), bitumen (i.e., oil sands), and kerogen (i.e., a waxy yellow lipid which is often found in shale). Petroleum gases are mainly comprised of methane, and include other gases such as ethane, propane, and butane. Coal is a similar type of organic hydrocarbon, but is not considered petroleum due the different circumstances under which it was formed and found.
Figure 2: Crude Oil Benchmarks by Density and Sulfur Content
Source: Refining 101: Basics of Refining and Coking
Formation of Hydrocarbons
Two theories explain the formation of petroleum resources.
The biogenic genesis theorem postulates that petroleum was formed hundreds of millions of years ago by deposits of organic material, which, through time, heat, and pressure became kerogen. Chemical processes which occur in the extreme conditions within the Earth’s crust then transformed kerogen into petroleum liquids and gases. Currently, there is limited economic rationale for commercializing kerogen, but it could be an important future source for hydrocarbons.
Over time, petroleum migrated from organics-rich source rocks (e.g., shale) into discrete accumulations called reservoirs or has remained in continuous deposits which pervade larger areas. The consensus of the scientific community is that the vast majority of recoverable petroleum was formed by biogenic synthesis. Due to unique conditions and the length of time which are required of biogenic processes, petroleum is seen as a limited and non-renewable (i.e., depleting) natural resource.
Figure 3: Biogenic Formation of Petroleum
= 180 million years ago today
Source: Al Hajeri, Saeed, et al. Basin and Petroleum System Modeling. Oil Field Review, 2009: 21, no. 2
An alternate theory, abiogenic genesis, suggests that at least some petroleum was formed in the absence of organic processes, through pyrolytic forces which occur deep within the Earth’s mantle. The abiogenic theorem has been largely been relegated to fringe corners of geology since it was postulated, but recent evidence indicates that a significant portion of natural gas resources, particularly gas hydrates (i.e., clathrates), lack organic “markers” (i.e., were not formed by biogenic synthesis)1. Methane clathrate has been demonstrated to exist in substantial quantities on ocean floors and could become an important future source for hydrocarbons, but the extraction of which has not been proven economic. Whereas the biogenic theory postulates that shales are the primary source rock for petroleum, proponents of abiogenic genesis posit that petroleum also forms in olivine rock formations via serpentinization (i.e., a natural analog of the Fischer-Tropsch process)2.
Figure 4: Abiogenic Genesis Theorem
Source: Kutcherov. Abiogenic Deep Origin of Hydrocarbons and Oil and Gas Deposits Formation. 16 Jan 2013
The Petroleum System: Petroleum Geology
A number of independent process influence the accumulation of hydrocarbons and the formation of petroleum systems (e.g., generation, migration, trapping mechanism, etc…). Geological, hydrostatic, and others factors determine the potential and commercial viability of a given petroleum system.
Figure 5: Formation and Accumulation of a Petroleum SystemSource: Crude Oil and Natural Gas: From Source to Final Product5
A petroleum system forms under certain conditions in which source-rock (e.g., organic-rich shale) generates hydrocarbons under the right amounts of temperature and pressure. Hydrostatic and thermodynamic forces may then cause some of these nascent hydrocarbons to migrate through permeable rock layers where they are eventually trapped by and accumulate underneath impermeable geological structures known as traps. The majority of all conventional petroleum resources are discrete accumulations which form within traps. Traps are further classified as structural or stratigraphic in nature3. Structural traps are due to localized geological features such as folding and faulting; stratigraphic traps are caused when alternating rock layers exhibit varying degrees of permeability.
Figure 6: Types of Hydrocarbon TrapsSource: Hydrocarbon Traps. 12 May 2014
Geological properties, such as permeability, porosity, and saturation influence the recoverability of hyrdocarbons. Permeability describes the ease by which petroleum flows through geological features; porosity describes the amount gas and fluids a given geological feature can hold; and, saturation describes the amount of petroleum actually contained within porous rock structures. The most commercially desirable rock formations are highly porous, permeable, and saturated with petroleum. Geological formations which are substantially porous and saturated, but not permeable, can be artificially induced to produce a commercially viable petroleum system by artificially inducing permeability (e.g., via fracking or matrix acidization).
Hydrostatic properties, such as pressure and drive mechanism, also influence the economic potentials of those hydrocarbon. In 1856, French engineer Henry Darcy derived an expression for hydraulic flow which is known as Darcy’s Law. According to Darcy’s law, the flow rate is proportional rock permeability, area, fluid viscosity, and pressure gradient — i.e., the petroleum which can be initially recovered from a given reservoir is a function of naturally occurring energy (i.e., pressure gradient between the reservoir and the surface) via various types of drive mechanisms. The most common driven mechanisms are water drive, depletion drive, and solution (i.e., gas) drive. In petroleum systems where hydrostatic pressures gradient are zero (i.e., well pressure equals surface pressure), pressure can be artificially increased through pumping (i.e., sucker-rod lift) and/or fluid and gas injection.
Initially, all oil production from a given reservoir undergoes transient flow (i.e., flow due near well-bore permeability) during which well pressures remain constant at the initial reservoir pressure, which is characterized by high production rates and also high decline rates. When the flow reaches the reservoir’s boundaries or when it meets the flow boundary another well, transient flow transitions to boundary-dominated flow. Since reservoir pressures will drop during boundary dominated flow conditions, this period is characterized by hyperbolic (i.e., decreasing rate) declines in production. Boundary permeabilities between multiple flow boundaries explains how some leases can become depleted by wells from nearby lease. This idea is perhaps more plainly explained in the final scene of There Will Be Blood, in which our anti-hero protagonist Daniel Plainview explains the ramifications of boundary-dominated flow to his estranged son-in-law before bashing his skull in with a bowling pin:
DRAAAIIINNNNAGE! Drainage, Eli, you boy. Drained dry, I’m so sorry. Here: if you have a milkshake… and I have a milkshake… and I have a straw; there it is, that’s the straw, see? Watch it. My straw reaches across the room… and starts to drink your milkshake: I… drink… your… milkshake![slurps]I drink it up!
– Daniel Plainview. There Will Be Blood.
Figure 7: I Drink Your Milkshake
Source: Know Your Meme. I Drink Your Milkshake.
Volume and depth are other important considerations for assessing the commercial and economic potential of petroleum systems.
The Petroleum System: Petroleum Geography
An idealized petroleum system has defined geographic, stratigraphic, and temporal boundaries. These boundaries exist in multiple dimensions: lateral (on the surface); vertical (i.e., depth), and time. This concept is perhaps best demonstrated by example — in this case, the Permian Basin of West Texas.
Looking at the Earth’s surface, petroleum systems accumulate in sedimentary basins. A named basin may also be subdivided into multiple discrete sub-basins. For example, the Permian Basin consists of three sub-basins, each with related (buy still unique) geology and stratigraphic taxonomy.
Figure 8: Permian Basin Geographic Map
Source: Murchison Oil & Gas Inc. Permian Basin.
Within a given sedimentary basin, there are often multiple plays (i.e., hydrocarbon producing areas within a basin) which correspond to specific geological formations underneath the Earth’s surface. Play names combine the names of the source rock and the major reservoir rock. The Permian Basin is host to several plays.
Figure 9: Permian Basin Play Geography
Source: Eric Roach. Inside DrilIinginfo’s Map Drawers #1: Permian Basin. DrillingInfo.com. February 13, 2014
If you were to flip the two-dimensional play map of the Permian Basin on its axis, you would get a glimpse of a vertical cross-section. A simplification of this cross-section is represented by a stratigraphic column chart.
Figure 10: Permian Basin Stratigraphy
Source: Murchison Oil & Gas Inc. Permian Basin.
Although, the stratigraphic column chart is an oversimplification of actual geological geometry, it is useful for identifying pay-zones (i.e., hydrocarbon producing intervals) within a given basin. The Permian Basin is host to several producing zones — each one its own play — and as a result is known as a “stacked pay area”. Recently, the Bone Spring, Wolfcamp/Wolfberry zones have attracted significant horizontal activity. The Montney Basin, located on Northwestern end of the prolific West Canadian Sedimentary Basin, also possesses significant stacked pay potential.
Stratigraphic charts which are applied to specific location (vice generalized over a large area) provide greater detail about uplift, fracturing, and other geological forces which are indicators for assessing the formation, migration, and accumulation of biological hydrocarbons. The below stratigraphic column chart comes from Apache Corp’s recent discovery of the Alpine High, a conventional hydrocarbon deposit which was lurking in a hotbed of unconventional activity.
Figure 11: Area-Specific Stratigraphy of the Alpine High Discovery in the Delaware Basin
source: Apache Corp. Investor Update of the Alpine High Discovery. Slide 13. 9 Oct 2017
Adding the time dimension to a point-location (cross-section of geography and depth) results in a burial chart. Burial charts fuse estimates regarding how organic sedimentation was influenced by depositional forces over time (i.e., temperature, pressure, compaction, uplift, erosion, and natural fracturing) to favor (or disfavor) the formation of hydrocarbons. In doing so, the burial history a given area suggests critical moments in time (a moment in geological time is a a few hundred thousand to a few million or so years) during which the formation of hydrocarbons was favorable. These critical moments, in turn, suggest which stratigraphic zones beneath the surface are likely to be the most hydrocarbon rich. Burial charts are mentioned extensively in the literature4, but examples (like this one, of the M G Nevill lease near El Paso) are hard to find, likely due to the proprietary nature of the information.
Type logs are a culmination of aforesaid data points and assessments providing a fusion of information on a specific area and/or specific well. Type logs fuse information about known stratigraphy and empirical observations from lithography (e.g., well core samples), minerology (e.g., vitrinite reflectance) and seismology. The resulting product has traditionally served as a petroleum geologists’ primary tool for assessing the commercial and economic potential of given location.
Figure 12: Well Type Logs from the Alpine High Discovery
source: Apache Corp. Investor Update of the Alpine High Discovery. Slide 13. 9 Oct 2017
Petroleum geologists further classify resources as either conventional or unconventional, according to the means by which they may be extracted and commercialized. Conventional resources are those which have petroleum producers have traditionally sought out due to the relative ease of recovery and commercialization using established techniques and technologies. Unconventional resources are those which require the application of unconventional extraction and processing techniques and technologies to bring to market. When unconventional technologies and evaluation techniques become standard practice, unconventional resources will be classified as conventional5.
Typically, conventional resources are accumulated within discrete geological traps. The discrete nature of these accumulations favors the use of vertical wells. Initially, production from conventional reservoirs is characterized by transient flow, but quickly –within a matter of days usually — transitions to boundary-dominated flow
Unconventional resources tend to exist in continuous (versus discrete) accumulations throughout large areas and, for this reason, are often called “continuous-type deposits”. Commercially viable unconventional resources include tight oil and gas (e.g., shale), coal bed methane, and bitumen (i.e., tar sands). Shales, for example, tend to be rich in petroleum but exhibit permeabilities below those which are required for a conventional reservoir system (i.e., below 0.1 millidarcies)6. Unlike within conventional reservoirs, transient flow from tight oil and gas can last a very long time due to the low permeabilities of surrounding rocks. This characteristic explains why shale wells decline very rapidly in the first few years, before transitioning to boundary-dominated flow.
In recent years, specialized techniques that artificially increase permeability have enabled widespread commercial recovery of petroleum within shale deposits. Due to advances in our understanding of geology and technology, tight oil and gas is probably no longer unconventional in the literal sense. Other unconventional resources, such as bitumen, may be readily recovered but require significant upgrading before they are commercialized.
Figure 13: Categories of Petroleum ResourcesSource: Cook. Petroleum Systems and Geologic Assessment of Oil and Gas in the Uinta-Piceance Province, Utah and Colorado. Chapter 23: Calculation of Estimated Ultimate Recovery (EUR) for Wells in Continuous Type Oil and Gas Accumulations of the Uinta-Piceance Province. 2003
The Fundamentals of Petroleum Recovery
Reservoir managers categorize phases of production according to the mean by which petroleum is extracted. The amount of petroleum which a reservoir originally produces through natural occurring energy within the reservoir — for instance, via gas drive or water drive mechanisms (i.e., without external stimulation) — is referred to as primary recovery. Secondary recovery methods utilize injection of water and/or gas to maintain hydrostatic pressure. Tertiary recovery methods, also called enhanced recovery (ER), include “injection of polymer solutions, surfactants, microbes, nitrogen or carbon dioxide, capable of influencing rock and fluid properties”7.
Generally speaking, the technical difficulty and expense rises as resources become depleted and as reservoir pressure drops.
Eventually, all wells deplete to the point of their ultimate economic limit. The point at which a well is no longer economically and/or commercially viable is ideally when estimated fixed plus variable production costs exceed revenues. The ultimate recovery of OPIP is extremely situational dependent, being governed by: types of petroleum resources in place; geological forces which dictate reservoir formation and flows; technology employed; and market conditions.
The Fundamentals of Resource Estimation: Recoverable Resources
Resource estimates allow planners and investors to quantify the recoverable portion of OPIP and determine production characteristics with respect to time. These estimates are key inputs to key economic metrics, such as present value calculations. The primary methods of estimating recoverable resources are volumetric, analogy, material balance, and production history. In the early stages of commercialization, where little empirical data is available, estimates will rely more heavily on volumetric analyses and analogy. As initial coring and log data becomes available, methods will shift towards the material balance method. Once a given area reaches commercial maturity, estimates tend to rely more heavily on historical production data.
Estimated ultimate recovery (EUR) is an estimate of the amount of OPIP which can feasibly will have been commercialized under various market and technological assumptions8. The ratio of EUR to OPIP, expressed as a percentage, is known as the recovery factor.
Reserves are specialized estimates of EUR which are valid only for a very specific set of assumptions. Although reserves are the most commonly reported and used resource-based metric for corporate valuation and lending, there is no universally accepted standard for what constitutes a petroleum reserve. According to the Independent Petroleum Association of America (IPAA):
…many entities have their own definition of “reserves”–the U.S. Geological Survey, American Association of Petroleum Geologists, Society of Petroleum Engineers, the Society of Petroleum Evaluation Engineers, the World Petroleum Congress, not to mention the Securities and Exchange Commission, the Internal Revenue Service, and the United Nations Framework Classification.
Although the Society of Petroleum Engineers (SPE) boldly attempted to reconcile the vast array of resource definitions9, any apple-to-apples comparison of internationally reported reserve numbers is a borderline exercise in futility. One could even say that global oil reserve estimates are a work of fiction.
Of note, judging from figures 15 and 16, below, proved reserves (1P; P90) offer a small glimpse of overall resource potential.
Figure 16: The Resource Pyramid
Even the point estimate for proved plus probable reserves (1P+2P; P50) is likely to underestimate the mean expected value for total resource potential. The differences between most likely and mean arise due to the fact that the distributions which underlie resource estimates are usually not symmetrical (vis-à-vis the normal distribution); rather, they tend to be right-skewed (vis-à-vis the lognormal distribution)10 11.
Figure 17: Probabilistic Resources Categorization
Source: Imre Szilágyi. The Value Of Petroleum Resources: Analyses Of Cash-Flows And Uncertainties. OGFJ. 11 July 2013.
Moreover, as a given reservoir or play matures, the uncertainty of resource estimates diminishes. As a result, estimates of proved reserves for commercially mature and well explored plays and fields will be far more indicative of the best estimate than for younger and underdeveloped areas. Yet, this characteristic of decreasing uncertainty is often exploited (manipulated?) to show investors proved reserve growth even in the absence of actual production or resource growth12. This “look at our proved reserve growth” gimmick is in addition to several other games that oil and gas producers play on investors.
Figure 18: Resource Estimate Uncertainty
Source: USGS. Arctic National Wildlife Refuge, 1002 Area, Petroleum Assessment, 1998, Including Economic Analysis. Fact Sheet 0028–01: Online Report.
Note to Figure 18: Schematic graph illustrating petroleum volumes and probabilities. Curves represent categories of oil in assessment. An example of how one reads this graph is illustrated by the blue and orange lines projected to the curve for economically recoverable oil. There is a 95-percent chance (i.e., probability, F95) of at least volume V1 of economically recoverable oil, and there is a 5-percent chance (F05) of at least volume V2 of economically recoverable oil.
Empirical research amplifies the obsolescence of reserve based value metrics. For example, the standardized measure of oil and gas is an estimate of the net present value of a company’s proved reserves — publicly traded oil and gas companies must disclose this measure annually as per the SPE13, FASB14 15, and SEC16 17 18. The standardized measure is basically an after-tax PV-10. One might think that this would be a powerful metric since it forces companies to value their disparate resource bases according to a common standard (e.g,. using the same resource metric; using a 10% annual discount rate; using the trailing twelve months’ average commodity prices; considering income taxes; etc…). However, the standardized measure is far from an economic measure since only “direct” costs need to be considered, and proved reserves are not necessarily indicative of the mean expected value for developed and undeveloped resource potential. Research from Capital IQ corroborates that this and basically all oil and gas industry-specific metrics are vastly inferior to traditional value metrics, such as price to cash flow, in predicting future equity returns19.
Suffice it to say, investors should refrain from relying heavily on narrowly defined and economically meaningless terms, such as reported reserves, even if we feel compelled to since these are precisely what companies disclose. If anything, all evidence points to the fact that estimates of recoverable resource quantities are products of economic estimates (a-la discounted cash flow analyses), and not the other way around.
The Fundamentals of Resource Estimation: Decline Curve Analysis
The timing of recovery (i.e., production) is a key factor of the present value of a given quantity of resource. The field of modeling which approximates the characteristic of production with respect to time is known as decline curve analysis. Its name is given by the universal observation that well, field, play, basin, and maybe even global production declines after an initial build-up and plateau. More advanced simulations of reservoir behavior, known as type curve analyses, are the domain of petroleum reservoir engineers.
Figure 19: Stylized production behavior of an oilfield
Source: Höök M, Söderbergh B, Jakobsson K, Aleklett K. The evolution of giant oil field production behaviour. Nat. Resour. Res. 18, 39–56. 2009
It should be noted that for most oilfields, economic thresholds occur on the well level, vice the field level. The result of variable well level decline rates has resulted in an abundance of stripper wells located throughout the United States which in total contribute about 10% of the nation’s crude oil and natural production. The major exception is when a single well or single grouping of wells, represents a singular project, such as is the case with offshore platforms. In this latter case, the stylized scenario depicted in figure 19 tends to be very close to the truth.
The primary intuition behind decline curve analysis is given by Darcy’s Law — that “the flow rate is proportional to the existing pressure gradient in the reservoir”20. So, as fluids flow from the reservoir to the surface, the pressure gradient drops, and therefore production can be expected to decline. Stated differently, the rate of production decline is proportional to the rate of reservoir depletion — under idealized conditions of exponential decline (e.g., constant flowing pressure of an incompressible fluid within a closed reservoir), the depletion and decline rates are, in fact, equal.
Although, it had previously been shown that exponential (i.e., constant rate) decline was the analytical solution to the flow from a well with constant bottom-hole flowing pressure21, in 1945 J.J. Arps applied this intuition to show that the loss (i.e., depletion) ratio was empirically correlated to real world production data. The subsequent exponential, hyperbolic, and harmonic equations Arps used are still collectively known as Arps’ Equations and are still used ubiquitously in reservoir and production modelling. Moreover, it has since been shown that exponential decline solves the diffusivity equation for a single phase, incompressible fluid flowing from a closed, pressure constant reservoir22 23. The connection between empirically-derived models and nature is robust.
For more in-depth discourse on decline curve analysis, I suggest:
- IHS’s (formerly Fekete’s) Harmony software guide which provides a concise overview of decline curve analysis.
- Dennis Coyne’s convolution methodology for reconciling the production and decline characteristics of individuals wells with those of their host fields.
Belaboring the geological factors which affect the economics of petroleum recovery eventually culminates in a single purpose: allocate capital in instances where price and value are dislocated. In publicly traded securities, price is easy to find. But in order to estimate value, investors have to make the best use of the information which is available to them. State and federal sourced data-sets often aggregate company-level data, so while they are often useful in determining basin or play economics, they do not reveal much regarding a company’s positioning within a given play. Commercial data-sets which can facilitate this analysis are often cost prohibitive. Therefore, the primary source of investor information is usually found within company disclosures, which are full of accounting jargon, boilerplate information, and legalese. A grasp of economic geology and petroleum resource management facilitates the proper interpretation of company disclosures — ultimately bringing investors closer to their goal.
Footnotes [ + ]