Natural Gas: Economics and Environment: A Handbook for Students of the Natural Gas Industry

By Harald Osel

Natural Gas: Economics and Environment is the fourth and last book in a series of textbooks that have focused on the technical and market related fundamentals of natural gas, while also taking aspects into consideration that one shared between the oil & gas industry.

Volume 4 consists of three additional chapters and commences w

• Language: English
• Publisher: Aurora House
• Release date: Apr 29, 2019
• ISBN: 9780648226239

Book preview

17

Natural Gas: Economics

The purpose of this chapter is to introduce the principles of calculating the economics of various natural gas projects and the economic factors influencing results. The various models to be discussed are the E&P model, the CCGT model and the LNG model. The E&P (exploration and production) model focuses on the economic conditions of an upstream gas project. The sales gas produced via an E&P project could subsequently be used as feed gas for a combined cycle gas turbine-based power plant, which is addressed in the CCGT (combined cycle gas turbine) model. The concept of liquefying natural gas and the economics associated with the process are further introduced in the LNG (liquefied natural gas) model. Factors of economic influence — such as the financing of a project, contractual background, and project-risks — are going to be covered via separate sections. A general description of the technical background for these models was provided earlier in section 15.2. (Volume 3) concerning combined cycle power generation (CCPG), and Chapter 12 (Volume 2) concerning the liquefied natural gas chain, or Chapters 4 to 11 of Volume 1, which deals with the various aspects associated with the E&P model.

The E&P-, CCGT- and LNG model take essentially four sets of input data into consideration:

(1) Quantitative data (e.g. reserves and production in case of the E&P model)

(2) Cost data with respect to investment- (CAPEX) and operating costs (OPEX)

(3) Market data (concerning sales quantities and prices)

(4) Government take (such as taxes or cost- and profit oil / gas split)

On the basis of this set of input-data profiles, the E&P model aims at evaluating an exploration and development project in terms of its profitability, measured by its ability to provide an annual cash flow over the lifetime of the project.¹ The CCGT model otherwise aims at providing economic calculations for a power generation plant, while the LNG model focuses on the economic production for a liquefied natural gas project with respect to the various elements of the LNG chain.

17.1. E&P Model and Information

The aim of this section is to provide an overview of the input data used for the E&P model. This includes specific data (such as reserves and production) as well as a general overview of topics including financing or risk management. Some issues discussed in this context — e.g. calculation methods — are valid not only for the E&P model but all other economic models as well. At the end of this section the details concerning calculations in connection with the E&P model are ultimately presented.

17.1.1 Method of Calculation and Sensitivity of Results

The applied method of calculation is based on standard techniques for the calculation of key economic indicators, especially the net present value (NPV) and the internal rate of return (IRR). Both NPV and IRR can be derived from the discounted project cash flow.

The net present value (NPV) can be defined as the discounted sum of each annual net cash flow (CF) over a period of time and is based on the following mathematical relationship:

The table below could further explain the NPV calculations in the formula above via an example where row (C) refers to the annual net cash flow as the sum of incoming (A) and outgoing (B) cash flows.

In this example, the undiscounted sum (D) of the annual net cash flows over 10 years is 100 monetary units, while it is only 21 in row (G) when discounted with a rate (i) of 10%. This difference reflects the time value of money as highlighted in row (E) where expenditures or earnings generated during later years are weighted lower than earnings or expenditures during earlier years. This can also be seen when comparing the values in row (C) and (F). The sum of all amounts in row (F) ultimately provides the NPV in row (G) of 21, while (F) = (C) * (E).

In the above example, the discount rate, i, is 10%, while the number of periods, n, is 10 with the time or year of the cash flow, t, ranging from 1 to 10 (t=0→n). The discount rate reflects the costs of capital and could be seen as an opportunity cost in the case that funds would not be invested in this project but, for example, in an equally risky investment in the financial market.

In other words: If the investor can’t gain a positive NPV @ 10% from the project he could opt not to invest but to put his money instead into a bank or to choose other types of investment in the financial market. What becomes evident when comparing rows (G), (H) and (I) is that the higher the discount rate, the lower the NPV, since the capital employed becomes increasingly expensive. The maximum interest- or discount rate that can be carried by the cash flow of the project without producing a negative NPV is the internal rate of return, IRR.

In the above example, the investment into the project breaks even at an interest rate of 15%. When defining the discount rate as the hurdle rate — i.e. the rate of return expected for a company’s investment — further attention could be given to calculating the prices (or other key project variables) required to solve the equation: NPV=0.

In the E&P, and all other models, further attention is given to calculating the sales price for natural gas to solve the equation NPV=0. The gas price derived in this way represents the ultimate output figure of the model and gives information about the minimum price required for a project. Comparing this price with the prevailing economic environment in the relevant market enables the economist to draw a conclusion on project feasibility. At the point NPV=0 the discount rate (which could also be viewed as the hurdle rate of the E&P company) is equal to the internal rate of return.

The variables of the various economic models are viewed, in this context, as being deterministic. A variety of more sophisticated calculation procedures could further be introduced to the model. Variables could for example be viewed with respect to their probability distribution (stochastic variables) and during more recent years, real option techniques have additionally been incorporated when calculating project economics.²

When considering variables to be deterministic instead of stochastic the model is also called unrisked. An unrisked model reflects decision making under certainty, which means that no probabilities have been attributed to, for example, the exploration risk and various other risks associated with an E&P project.³

In case all possible results of a decision making process could be weighted with a specific probability, the average of all the results is referred to as the expected value or expected monetary value (EMT) while referring to figures in monetary terms. The schematic expression of expected value computations could be visualized in the form of a decision tree⁴ consisting of nodes connected to branches, all of which represent alternatives during the decision making process. At every node of the decision tree below an alternative route exists concerning decision making, while a probability of x% and y% could be attributed to the probability of each outcome (with 0 ≤ x ≤ 1 and 0 ≤ y ≤ 1 and x + y ≤1) .

The decision tree aims at calculating the expected value of alternative decisions by nodal analysis, i.e. by analysis of the decisions at various nodes. When statistical, random sampling methods are used to approximate an expected value, the method is known as simulation.⁵ During the evaluation of the results of the economic models presented in the following, no decision trees or simulation methods are going to be used.

Various scenarios could alternatively be considered in connection with a sensitivity analysis by altering the values of key input parameters to evaluate the sensitivity of output variables with the base case corresponding to 100%. The spider diagram below shows the effect of a percentage change of the various input parameters (vis-à-vis the base case) on the net present value at a discount rate of 10%, NPV(10).

The following diagram shows the effect of the same sensitivity calculations but on the internal rate of return, IRR (instead of the NPV).

The base case (100%) can be seen as the most likely case where best estimates have been used to define the time-series of the input parameters such as capital expenditures (CAPEX), operating expenditures (OPEX), prices, inflation or production. (Over the life of field, production is equivalent to the recoverable reserves.) Economics could be calculated by using nominal values including inflation or real values where the effects of inflation have been removed. Values can be converted from nominal terms to real terms by adjusting for inflation (i.e. inflation adjustment).

17.1.2 Economics and Accounting

All models (E&P, CCGT, LNG) exclusively focus on economic values that could be derived from the cash flow generated by the project (cash earnings).

The annual financial reports of an oil and gas company are not focused on cash flow values but on earnings, i.e. the accounting profit, which is also referred to as net earnings or net income.⁶ The accounting profit represents the realized profit following the deduction of all costs and expenses — as well as all taxes and interest — from the income generated by the company through its sales.⁷ In the case the financial result (esp. taxes and interest) would still remain excluded, the corresponding financial performance figure is referred to as the EBIT, i.e. earnings before interest and taxes and reflects the operating result of the company. For an upstream oil and gas (E&P) company the gross income further corresponds to the revenues generated from hydrocarbons produced at or near the wellhead, prior to the deduction of costs, depletion, royalties, taxes and interest.

The difference between the operating cash flow and the accounting profit are the non-cash expenses, primarily all DD&A, i.e. depreciation, depletion and amortization,⁸ as well as the accrued values.⁹ For project evaluation purposes it is the net operating cash flow after tax that is of primary concern. (EBITD, i.e. earnings before interest, taxes and depreciation corresponds roughly to the annual operating cash flow, especially in case of low accruals.)

There remains to be a borderline where the two approaches (Economics and Accounting) do meet. For example:

(1) When having to forecast values required for accounting (such as for the calculation of decommissioning provisions), project economics based on the company’s best estimate need to be used.

(2) When calculating what is called in US GAAP, a ceiling test and according to IFRS, an impairment test,¹⁰ the NPV of the projects of the company (in the sense of cash generating units) is compared with the net book values (NBV) or net carrying values (NCV). In this context, accounting rules provide for a special depreciation of the fixed assets to the extent of NPV < NBV. That is, if the future income in the form of the NPV is less than the NBV of the asset, the difference needs to be impaired. (The NBV is the original, historic value of a fixed asset minus the annual depreciation for the years in use.)

The perspective chosen when running an economic model is, in most cases, a single project. Accounting related data focuses on all projects at a company level.

17.1.3 Reservoir: Natural Gas Reserves and Production

All efforts made by an E&P company are ultimately directed towards a single goal, which is the commercial production from a hydrocarbon reservoir.

17.1.3.1. Natural Gas: Definition

A summary of terms used in the oil and gas industry in connection with natural gas (NG) can be found in schematic below.¹¹

It is worthwhile to note that the notion of natural gas is not only used to refer to methane natural gas (usually with some ethane or even higher hydrocarbons contained within) but also to natural gas in a gas reservoir. Natural gas contained within a hydrocarbon reservoir remains a fluid that is comingled with other fluids. That fluid-mix is produced as raw gas, which is a mixture of various hydrocarbon gases (HCs) and gas contaminants (such as CO2 or H2S). Once raw gas has been brought to surface and the gas contaminants have been removed in a treatment plant, the higher hydrocarbons can be removed to gain butane, propane or condensate. After gas treatment and -processing (see section 9.) the remaining hydrocarbon gas is essentially methane (usually with some ethane) and could be referred to as natural gas of a sales gas quality.

Raw gas is a notion that includes gas contaminants in addition to hydrocarbon gases.

The term natural gas is in fact used to refer to hydrocarbon gases in a reservoir (other than condensate), while it also refers to methane natural gas when shipped as sales gas from the outlet of the gas plant.

For a gas reservoir, the term hydrocarbon gas refers to C1, C2, C3, C4 being present in the formation as a gas mixture at reservoir conditions. Even though C5+ is also a gas at typical reservoir conditions (concerning pressure and temperature), it is referred to specifically as condensate. (Only in a gas-condensate reservoir below dew point does C5+ condense as a liquid.)

Natural gas liquids (NGL) and condensate could be gained as liquids on surface either via gas processing, such as fractionation (in case of C2, C3, C4), or due to condensation at ambient pressure and temperature (in case of C5+).

Following fractionation in the gas plant, propane (C3) and butane (C4) can be liquefied in a LPG plant (also bottling plant) in order to be kept in pressurized gas-bottles.

Methane natural gas could ultimately be liquefied in a LNG plant also and marketed as liquefied natural gas (LNG).

17.1.3.2 Reserves

Economic modelling of the size of hydrocarbon reserves needs to take place at an early stage. At this point only some seismic data and a general geological concept of the area might be available. From this information, the input data set for the economic model needs to be derived including potential reserves and the corresponding production profiles.¹²

The quality of the economic model could subsequently improve in tandem with an increase of information available during the progress of the project. This is the case, for example, when geological data is supplemented by data from well-logs or geophysical data gained from an interpreted 3D seismic, especially when PVT (pressure-volume-temperature) data or well testing data becomes available. This additional data can be used to create a reservoir model, which could provide an improved understanding of the amount of oil and gas originally in place. Reservoir modelling at the point of discovery must necessarily remain incomplete due to a lack of reliable data. During later stages and ongoing production, the information gathered about the reservoir parameters continues to be upgraded. Following the commencement of production, the reservoir data could be updated with data gained from reservoir simulations based on the reservoir model¹³ as well as data derived from a statistical approach towards reservoir modelling.¹⁴ Modern reservoir models are otherwise computer-supported, three-dimensional (3D) reservoir models.

The goal of model-building remains to be the simulation or forecast of key reservoir parameters, such as the original quantity of hydrocarbons in place or future production profiles. Reservoir modelling further targets facilitating decisions concerning issues such as determining the drilling-location of wells, optimizing the set-up for surface facilities, or supplying information required for secondary- or enhanced recovery methods.

At the beginning of an E&P project, be it a natural gas or oil project, the evaluation of reserves could rely only on volumetric methods based on seismic studies. Subsequently, data acquired during drilling operations could be used as well. Only following the production of about 5-10% of the recoverable reserves¹⁵, might sufficient data become available to enable the reservoir engineer to incorporate more accurate production forecasts into the reservoir simulation studies, such as material balance studies (see 4.3.6.2.) and the analysis of decline curves (which are a plot of production rates versus time).

Petroleum Reserves — Definitions: Petroleum quantities from known hydrocarbon accumulations, which could be commercially recoverable, are referred to as reserves.¹⁶ The degree of recoverability together with the estimate of petroleum quantities held within the reservoir cannot be predicted with certainty. It was consequently found useful to further classify reserves as proved or unproved reserves, according to their expected probability of recovery.

Proved reserves are reserves that could be commercially recovered with a reasonable degree of certainty. Proved reserves are sometimes also referred to as P90 reserve, which means that with a probability ≥90% originally estimated reserves correspond to actually recoverable reserves. Proved reserves could further be classified as proved developed or proved undeveloped depending on the status of reservoir-development and available marketing options.

Unproved reserves are reserves where commercial recovery could not be achieved with a high degree of certainty. Unproved reserves could be classified according to two subcategories, which are the probable reserves and the possible reserves. Probable reserves are also referred to as P50 reserve, meaning that with a probability ≥50% originally estimated reserves and actually recoverable reserves coincide. Similarly, possible reserves are reserves, which could be recovered only with a probability of ≥10%.

17.1.3.3 Production

The definition of reserves necessarily overlaps with the definition of production as highlighted in connection with wellhead production, marketable production and sales volumes in the following schematic.

(Source: OMV)

The production of natural gas or hydrocarbons is not taking place at constant production rates.¹⁷ Three successive phases could be differentiated as the:

(1) Start up or ramp up period, in which production commences at the beginning of field development. In tandem with more development wells coming on, stream production increases until development was finalized and stable production rates could be established during

(2) Plateau production. During this period, stable production rates could be monitored at peak levels until

(3) Decline sets in at the end of peak production. During decline, production rates gradually decrease in the tail phase of production until depletion of the field.

The figure below shows a production profile that summarizes these three phases schematically. Overall, a steady decline could be monitored vis-à-vis the absolute open flow (AOF) potential identified during deliverability testing (see 7.3.3.3.) carried out prior to the commencement of commercial production.

This plot of rates of production from a reservoir versus time is known as a production profile. Estimated profiles form part of production forecasting even to the end of field-life.

Reservoir evaluation models and production forecasting methods remain to be interlinked and could be differentiated according to four general types of models used for estimating reserves and forecasting production, i.e.:

(1) volumetric methods

(2) material balance calculations

(3) reservoir simulation and

(4) decline curve analysis

Volumetric methods make use of static reservoir properties or properties of the reservoir rock derived from sources such as log- and core data gained during drilling. A volumetric analysis aims at calculating the amount of oil- or gas in place by combining the above mentioned data with geological maps (e.g. structural- and isopach maps). Pore volume data derived from rocks and geological data concerning reservoir area and thickness could be combined to calculate a volumetric estimate for the oil- or gas in place. Volumetric calculations make use of static properties and consequently do not allow for the forecasting of reservoir performance in connection with dynamic data gained during testing and production.

Volumetric methods are generally used at a very early stage of reservoir performance, even prior to production from the reservoir.

Material balance calculations make use of pressure-related reservoir rock- and fluid properties and incorporate them into a single tank reservoir-model together with the production history of the field (while assuming steady state flow conditions).

Reservoir simulation on the other hand is based on a system of 2D or 3D grids that form the basis for gaining a variety of field data, which could be further used as input data for flow equations (diffusivity equations), equations of state and material balance equations. All these types of equations could ultimately be combined to form a single reservoir simulation model. The effects of production history on all these equations (in connection with specific cells within the grid) could further be incorporated into the model.

Decline curve analysis¹⁸ is built upon the basic concept of defining a production decline rate, D, which is calculated as the product of the change of flow rate over time, dq / dt, and is inversely related to the flow rate itself, -(1/q), i.e.:

D = — (1/q) * dq/dt

This formula refers to a curve describing the decline of the production rate over time. A plot of the actual production rate versus time could subsequently be fitted to a production decline curve in order to provide a tool for forecasting future production. From these forecasts, estimates concerning total reserves could be derived. Decline curves generally aim at forecasting production from an individual well. The decline curves of all wells within a field describe a set of production decline curves for the entire reservoir.

In the ideal case of availability of all data required, this concept could be extended to comprise of multiple field decline curves or even regional- or country specific curves until an overall, global decline curve for worldwide production could be derived.¹⁹

Field production profiles need to be differentiated from decline curves referring to production or production forecasts from a single well. Decline curves focus on production from an individual well linked to a decrease of its absolute open flow (AOF) potential since the start-up of production. A simplified decline curve was introduced in the figure above. (In reality, the hyperbolic-, exponential- and harmonic decline curves actually monitored do not correspond to a linear function such as in the above figure) Decline curves from one well could further be compared with the behavior during actual production from another well to add to the quality of production forecasting via decline curve analysis.

The entire economic life of a field is usually in the order of several decades (> 20 years). Examples for a short field life (e.g. in connection with condensate banking in a gas-condensate reservoir) or an unusually long production period²⁰ could further be presented. The total production period, as well as the length of the different phases in the figure above, largely depends on the geological characteristics of the reservoir. Every reservoir is, to a certain degree, unique in a geological sense and requires a specific approach towards its development.

An important factor influencing the recoverability from a reservoir is its drive mechanism. Discussions concerning reservoir drive mechanisms²¹ focus on natural drive energy, i.e. the mechanism prevailing within the reservoir that provides the pressure to move hydrocarbons to the surface (see section 2.2.1.1.). Expanding gas and inflowing water are the two main drive mechanisms. Gas drive (also called depletion drive) could either be provided by expanding gas coming out of a solution with oil (i.e. solution- or dissolved-gas drive) or by a gas cap expanding during the course of production (gas cap drive). Water drive is provided by water expanding below the hydrocarbon layer and pushing hydrocarbon fluids towards the wellbore.

A third mechanism — compaction drive — could be active in unconsolidated formations, which tend to collapse during continued production, thus increasing reservoir pressure, ideally without reducing permeability. A compaction drive mechanism could occur with all types of reservoir (holding oil, gas or gas-condensate). With many reservoirs a combination of different drive mechanisms could be at work, which is sometimes referred to as combination drive. All of these drive mechanisms are collectively referred to as primary drive.

Solution gas and gas cap drive (including segregation drive) are mechanisms that typically refer to oil production (involving the production of associated gas or casinghead gas from oil reservoirs).

Gas reservoirs must have some form of depletion drive where the expansion of the compressed gas in the reservoir is created by the pressure differential between high reservoir pressure and low wellbore pressure that drives gas towards the well and further towards the surface. The higher the reservoir pressures during field discovery, the higher the recovery efficiency that could be expected from the field. Highly pressurized dry gas reservoirs of a volumetric-type with a depletion drive and no movement of water into the producing formation (water encroachment) exhibit the highest recovery efficiency of all types of reservoir with recovery factors of up to 90%.

An active water drive mechanism could be found with all types of reservoirs. Although a water drive could be very efficient for the displacement of oil from porous media, it is less efficient with displacing gas. In comparison with a gas reservoir governed by depletion drive, a water-driven gas reservoir should be expected to be only half as efficient concerning recovery. The reason for this is high interfacial tension between water and hydrocarbon fluids in the gas reservoir. As a consequence, the attraction between water and the reservoir rock increases relative to the attraction between gas and rock. This results in water bypassing accumulations of gas, which creates a permeability barrier that hinders the further movement of such a gas pocket. One way to avoid this is to produce water from the aquifer simultaneously with gas from the formation to avoid an upward movement of the gas-water contact and water encroachment.

Depending on the kind of reservoir under consideration, the type of primary drive mechanism in place could contribute to recovery factors between 20% to 80% of the original hydrocarbons in place. High pressure dry gas fields²² should be, due to their superior flow characteristics, at the high end of this range, while oil fields (especially those with a solution gas drive) generally exhibit rather low primary recovery factors. Secondary recovery or enhanced recovery techniques in general could markedly enhance the overall recovery from such a field. In the case of a dry gas field, the primary drive mechanism might even be sufficient to achieve recovery rates of >80%. Indicators for recovery rates concerning the main types of drive mechanisms have been summarized below.²³

The first two types of drive mechanism refer to oil fields or liquid hydrocarbon reservoirs. A water drive could further be present in oil- as well as gas fields. An expansion drive that could lead to recovery rates of >80% could otherwise be associated only with a dry gas reservoir.

Solution Gas Drive: 5-30%

Gas Cap Drive: 20-40%

Water Drive: 35-75%

Dry Gas Field: >80%

Dry gas reservoirs with an expansion drive could achieve recovery factors as high as 80-90%. Installing compressors at the wellhead might be required at a later point when reservoir pressure drops, while a certain pressure level remains to be required for the shipment of gas to the flow lines. In the suction area of the compressor — connected to the tubing inside the well — a pressure drop can be achieved, which increases flow from the reservoir to the wellbore due to an increase in the drawdown. By installing wellhead compressors, the recovery factor of a volumetric gas reservoir could increase, in the ideal case, to >90%.

A retrograde gas-condensate reservoir of a volumetric type might achieve similar recovery factors of up to 80-90% but only when properly cycled. If this was not the case and condensate banking occurs in the reservoir, only a fraction of this recovery factor would actually be achievable (see 4.2.3.7.).

Although an aquifer could be regarded as a positive indicator in the case of an oil reservoir, a water drive causes, in addition to providing drive, the entrapment of gas, or gas-condensate pockets in the case of non-oil fields. The recovery factor of gas reservoirs with an aquifer and water drive would typically be only in the range of 60-80%.

A gas-condensate field with a water drive could achieve a similar recovery factor of about 60-80% in cases of cycling the gas cap first and producing the remaining dry gas only thereafter. In the case that the cap would not be cycled already at the beginning of production, which contributes to maintaining reservoir pressure, water from the underlying aquifer could flow into the reservoir and cause the entrapment of reservoir fluids. Consequently, gas reservoirs of a purely volumetric-type are producing higher recovery factors than those with an underlying aquifer.

As mentioned previously, production was assumed so far to result from the primary recovery in connection with a sufficiently permeable, homogenous reservoir, which corresponds to ideal conditions. In reality, additional production techniques linked to secondary- or enhanced recovery methods (see section 2.2.1.2.) need to be applied once production based on primary drive mechanisms starts declining.

Primary recovery refers mainly to the primary drive mechanism of the reservoir, while artificial lift technology is often applied in addition once reservoir pressure decreases due to depletion. Artificial lifting focuses on moving reservoir fluids to surface by using a variety of pumps, such as sucker rod pumps,²⁴ hydraulic pumps (including jet pumps), electric submersible pumps, as well as gas lift and plunger lift technology.²⁵

At this point it is worthwhile to