Brandt, A. R. October 12, 2011. Oil Depletion and the Energy Efficiency of Oil Production: The Case of California. Sustainability 2011, 3, 1833-1854

[ Brandt estimates the energy return on investment (EROI) dropped from ~6.5 in 1955 to ~3.5 fifty years later in 2005, mostly due to enhanced oil recovery, and the secondary effects of oil depletion.

California oil production reached its peak in 1984 at 1.2 Mbbl/d, and is in terminal decline [20]. Per-well yearly production rates are currently less than 5,000 bbl/well, down from a peak of about 24,000 bbl/well in 1930. Over 17% of California oil production in 2008 was produced from stripper wells—defined by the California Department of Oil, Gas, and Geothermal Resources as wells producing less than 10 barrels per well per day. Some 59% of operating wells in California are now classified as stripper wells [21].

Excerpts (out of order) from this 22 page paper follow. Figures/tables not included]

Abstract: This study explores the impact of oil depletion on the energetic efficiency of oil extraction and refining in California. These changes are measured using energy return ratios (such as the energy return on investment, or EROI). I construct a time- varying first-order process model of energy inputs and outputs of oil extraction. The model includes factors such as oil quality, reservoir depth, enhanced recovery techniques, and water cut. This model is populated with historical data for 306 California oil fields over a 50 year period. The model focuses on the effects of resource quality decline, while technical efficiencies are modeled simply. Results indicate that the energy intensity of oil extraction in California increased significantly from 1955 to 2005.

This resulted in a decline in the life-cycle EROI from˜6.5 to˜3.5 (measured as megajoules (MJ) delivered to final consumers per MJ primary energy invested in energy extraction, transport, and refining). Most of this decline in energy returns is due to increasing need for steam-based thermal enhanced oil recovery, with secondary effects due to conventional resource depletion (e.g., increased water cut).

As the quality factors declined in favorability in the California oil industry (e.g., more water must be lifted for each unit of oil produced), the increase in technical efficiencies did not fully compensate for these reductions in quality. This trend caused the energetic returns to oil extraction to decline significantly over the modeled time period.

A transition in oil production has been occurring for decades: the fuels that consumers put into their automobiles are being produced using increasingly energy-intensive production methods, and from resources other than “conventional” oil.

This oil transition will cause growing tension in the coming decades: a transition to low-quality oil resources will reduce our ability to improve the environmental profile of energy production—an imperative for the 21st century—but increasing demand for fuel from developing countries could increase market instability and competition over constrained oil resources.

This transition is the result of three trends occurring worldwide:

• Output from existing oil fields is declining
• New fields are not as large or productive as old fields
• Areas with conventional resources are increasingly off-limits to investment by independent oil companies.

These trends are inducing investment in substitutes for conventional petroleum, such as the Alberta tar sands, or synthetic fuels from coal or oil shale [1].

The most common substitutes for conventional oil have been low-quality hydrocarbons, such as the heavy oils in California and bitumen in Alberta. These resources are more difficult to extract than conventional petroleum, are more difficult to refine into finished fuels, and are more expensive. Much of this increased cost and difficulty is due to larger energy demands for extraction and refining.

For example, in California, thermally-produced heavy oil requires the injection of steam to decrease the oil viscosity and induce flow within the reservoir. Also, refining heavy oil is more energy intensive due to the fact that it is hydrogen deficient and often impurity-laden.

The nature of oil depletion is understood mostly by studying aggregate statistics such as regional production curves [2–4]. Due to the lack of publicly available data, little research has been performed on the specific effects of depletion on oil operations (e.g., effects of depletion on required capital investment versus operating expenses). Also, only a small amount of attention in the peer-reviewed literature has been paid to the energy efficiency impacts of oil depletion [5,6].

This paper seeks to explore these energy efficiency impacts by building a detailed model of California oil production over time. First this paper presents a history of California oil production, focusing on changing oil resource quality and resource depletion. Next, methods for calculating energy inputs and outputs from oil production are described. Using these energy inputs and outputs, energy return ratios are computed using methods of life cycle assessment (LCA) and net energy analysis (NEA). Lastly, results from these calculations are presented and their broader significance is discussed.

An Industrial History of California Oil Production: Resource Quality, Depletion, and Innovation

Early Oil Production Before 1900. Pre-commercial use of oil in California included use by Native Americans for coating, sealing and adhesion [7]. Early commercial production was concentrated in the San Joaquin valley of California, where low-quality surface oil was mined in pits and tunnels [8]. The first refinery—a simple still for batch processing with a capacity of 300 gallons—was constructed near McKittrick in 1866, but it soon failed due to poor economics and high transport costs [7].

By the 1890s, surface mining had declined in importance and the cable tool rig had become the standard drilling method in California. Power for the cable tool rig was provided by a steam engine, and drilling power increased rapidly: Drake’s famous rig generated 4.4 kilowatts (kW) in 1859, and by 1900, steam boilers were rated at 29 kW, and the attached steam engines were rated at 18 kW [7].

Oil became a socially and economically dominant industry in the San Joaquin Valley with the discovery of the Kern River field in May of 1899. The proximity of this field to Bakersfield allowed the shipment of oil via rail to San Francisco. By 1903, Kern River production increased to 17 million barrels per year, or˜ 70% of California’s production. No gushers were ever found in the Kern River field, due to its low initial pressure (1.2–3.8 megapascal (Mpa) and heavy viscous oil (0.96–1.0 specific gravity, and up to 10, 000 centipoise viscosity) [11]. Early oil production was inefficient and wasteful, due to a combination of poor knowledge of geological principles and poor ability to control production. Early producing wells often declined rapidly, particularly in the Los Angeles basin [7]. This is because producers would withdraw and often vent or flare the associated gas, depleting the reservoir drive. These depleted wells, generally producing only a few barrels per day, were often sold off by their operators. Industrious producers would buy up contiguous depleted wells and apply technology to increase production. Operators would commonly attach a central oil-fired pumping unit to serve numerous wells simultaneously [7]. This represents an early example of self- consumption of oil by producers to offset the effects of depletion.

Early 20th Century Production: 1900–1940. The first recorded attempt at thermal enhanced oil recovery (EOR) was by J.W. Goff in 1901 [ 7]. He injected steam, air, and steam-heated air into wells and achieved a small amount of incremental production, but depressed oil prices stymied his early attempt at EOR [7].

An important technological development in the early 1900s was the advent of the rotary drilling rig as a replacement for the cable-tool rig by Standard Oil in 1908. Early rotary rigs were powered by steam engines, which were replaced by internal combustion engines by the 1930s. Rotary drilling eventually dominated oil well drilling, as it was faster and more effective. Efforts in the late 1920s and 1930s focused most visibly on attempting to drill deeper. Deep oil had been found at Kettleman hills: 635 m3/d (4000 bbl/d) from a single well, 2133 m deep, producing valuable light oil (0.74 specific gravity) [12]. This encouraged others to drill deeper wells in existing fields. Thermal methods were also experimented with briefly in this period, with the Tidewater oil company injecting hot water into the Casmalia field in 1923 (Casmalia oil is dense and viscous, having a specific gravity as high as 1.015 [11]).

These early attempts at enhanced oil recovery were not successful, as ample production from high-quality light oil fields at the time made these operations costly and unneeded. Per-well yearly production rates peaked in the 1930s, reaching ˜ 24,000bbl/well in 1930 and declining thereafter to less than 5,000 bbl/well in the current day.

The Modern Era of California Oil Production: 1940 to 2000. In the post-war period, discovery of large new fields declined. Research attention focused on ways to extract a larger share of California’s vast heavy oil resources. Knowing that heat reduces the viscosity of crude oil, in 1956 engineers attempted to light a fire downhole in the Midway-Sunset field by injecting air and using a novel electric ignition system. This method is called in situ combustion or “fireflooding”. The ignition system was unnecessary, as injected air caused spontaneous combustion [13].

Other companies utilized bottom hole heaters. These heaters took heat generated at the surface and transmitted it to the formation using a heat exchanger. Engineers concluded that heat conduction from the bottom hole heaters was slow and ineffective, and that more effective thermal production would require injecting heat-conducting fluid into the reservoir body.

The first modern steam injection project recorded in California was in April of 1960. Many steam injection projects were built quickly: in 1964 and 1965 more than 50 steam injection projects were initiated each year [14]. Production increased significantly in fields where steam injection was instituted.

Oil production continued to increase in the 1960s. Production increased to over 1 Mbbl/d in the mid 1960s, reaching a plateau that lasted about 20 years [15]. Simultaneously, production per well dropped, reaching 25 bbl/well-d in 1963 and never rising above this level again [15]. This is because much of the incremental production in this period was not from new large fields or gushers, but instead from increasing the intensity of extraction in depleted fields using advanced recovery technologies.

A Bottom-Up Model of Energy Inputs and Outputs. Our process model includes three process stages: primary energy extraction, upgrading of primary energy into forms usable by consumers (in this case refining), and consumption of refined energy in non-energy sectors. Both direct and indirect consumption of energy in oil extraction is accounted for as the flow of refined product back into the system (e.g., the model includes both refined fuels used directly in oil extraction, such as diesel fuel used in drill rigs, as well as those fuels consumed indirectly, like diesel fuel used during steel manufacture). This model formulation—with self-consumption included— accounts for the fact that a fraction of the primary energy produced is used to extract more primary energy.

Calculating Energy Return Ratios. Oil has been the subject of a number of NEA studies that have calculated energy return ratios [5,6,30–32], but previous analyses have generally been based on high-level datasets (e.g., national datasets). There are a number of energy return ratios used in NEA. Defined most simply, the net energy output from an energy extraction and refining process is the energy made available from a natural resource in useful, refined form less that energy consumed in extracting, upgrading and converting it to that form [24]. Energy return ratios of various types can be constructed, generally with a measure of energy output in the numerator and a measure of energy consumed in the denominator. The most common energy return ratio is the net energy ratio (NER), also called the energy return on energy invested (EROI) [33]. Other metrics include the external energy ratio (EER)

The EER compares energy inputs from outside the system to net outputs from the process. It reflects the ability of a process to increase energy supply to society. The NER compares all energy inputs to net outputs. It is therefore a better metric for understanding environmental impacts from producing a fuel (e.g., GHGs) [34]. There are two possible system boundary configurations when deriving EER: refined fuel consumed by the system itself can either be considered an internal or external energy source. For example, diesel fuel used to power drilling rigs could either be considered an internal energy source, (“loose” system boundary) or could be considered a final energy product that is diverted back into the process (“tight” system boundary). For the EER calculated here, the model uses the tight system boundary. This choice is made because the refined fuel leaving the refinery gate is used for final consumption, so its diversion back into oil extraction is classified as an external energy input.

A difficulty with this bottom-up modeling approach is that there is no clear way to separate the oil energy extraction chain completely from other extraction chains such as coal production. For example, some of the final products from oil and gas extraction will in fact go to other energy extraction sectors either directly or indirectly, not to non-energy end consumers. This is a related problem to the general system boundary problem in LCA: determining where your “system” begins and ends is not trivial and there is no unambiguously correct approach to doing so. These complexities are ignored for the first-order model created here.

Energy return ratios give insight into the quality of the resource: a high quality resource will require less energy to extract and upgrade than a low-quality resource. These ratios also give some sense of the efficiency with which industry is able to extract resources. Over time, as technologies become more efficient and their usage is systematically improved through research and development, the energy return ratios will improve for a given level of resource quality. Energy return ratios are only partially correlated with other metrics of interest, such as the cost of a resource and the its environmental impacts [24]. In their favor, however, they can illustrate fundamental qualities of the resource that can be obscured by economic or environmental metrics.

Clearly then, the energy requirements of crude oil extraction and refining depend both on the quality of the resource and the technical efficiency with which industry extracts and refines the resource. For example, quality factors might include the volumes of water lifted per unit of oil produced, or the depth of fields accessed over time. Efficiency factors might include the efficiency of pumps or the refining energy intensity.

The uncertainty in model results is significant. EROI values actually achieved in the California oil industry over time are fundamentally unobservable: many of the required data inputs are not publicly available or were likely even lost over time due to neglect. This lack of data causes fundamental difficulties in assessing the uncertainty.

References

1. Farrell, A.E.; Brandt, A.R. Risks of the oil transition. Environ. Res. Lett. 2006, 1, 014004.

2. Brandt, A.R. Testing hubbert. Energy Policy 2007, 35, 3074–3088.

3. Deffeyes, K.S. Hubbert’s Peak: The Impending World Oil Shortage; Princeton University Press: Princeton, NJ, USA, 2001; p. 208.

4. Campbell, C.J.; Laherrere, J. The end of cheap oil. Sci. Am. 1998, 278, 78–83.

5. Hall, C.A.S.; Cleveland, C.J. Petroleum drilling and production in the United States: Yield per effort and net energy analysis. Science 1981, 211, 576–579.

6. Cleveland, C.J. Net energy from the extraction of oil and gas in the United States. Energy 2005, 30, 769–782.

7. Rintoul, W. Spudding in: Recollections of Pioneer Days in the California Oil Fields; California Historical Society: San Francisco, CA, USA, 1976.

8. Rintoul, W. Drilling through Time: 75 Years with California’s Division of Oil and Gas; California Dept. of Conservation Division of Oil and Gas: Sacramento, CA, USA, 1990.

9. API. Petroleum Facts and Figures: Centennial Edition; American Petroleum Institute: New York, NY, USA, 1959.

10. API. Basic Petroleum Data Book: Petroleum Industry Statistics. Basic Petroleum Data Book; American Petroleum Institute: Washington, DC, USA, 2004; Volume 24.

11. CDC-DOGGR. California Oil and Gas Fields, Volumes I–III. Technical report. California Department of Conversation, Division of Oil, Gas, and Geothermal Resources: Sacramento, CA, USA, 1982–1998.

12. Rintoul, W. Oildorado: Boom Times on the West Side; Valley Publishers: Fresno, CA, USA, 1978.

13. Rintoul, W. Drilling Ahead: Tapping California’s Richest Oil Fields, 1st ed.; Valley Publishers: Santa Cruz, CA, USA, 1981.

14. CDC-DOGGR. Summary of Operations: California Oil Fields. Technical report. California Department of Conservation, Division of Oil and Gas, and Geothermal Resources: Sacramento, CA, USA, 1966.

15. CCCOGP. Annual Review of California Oil and Gas Production. Technical report. Conservation Committee of California Oil and Gas Producers: Bakersfield, CA, USA, 1994.

16. Dennison, W.J.; Taback, H.; Parker, N. Emissions Characteristics of Crude Oil Production Operations in California. Consultant report KVB72-5810-1309ES. California Air Resources Board: Sacramento, CA, USA, 1983.

17. Norton, J.F. A Report to the California Energy Commission: The Options for Increasing California Heavy Oil Production: Final Report; Radian Corporation: Sacramento, CA, USA, 1981

18. Henwood, M.I. Feasibility and Economics of Cogeneration in California’s Thermal Enhanced Oil Recovery Operations; California Energy Commission, Assessment Division: Sacramento, CA, USA, 1978. 19. CDC- DOGGR. 2003 Annual Report of the State Oil & Gas Supervisor. Technical report. California Department of Conservation, Division of Oil, Gas and Geothermal Resources: Sacramento, CA, USA, 2004.

19. Production is lightly taxed in California (e.g., no mineral severance tax), the reduction in California oil production is not likely a major driver of California’s recent budget difficulties. The production decline since peak production amounts to˜ 200 Mbbl/y, which if valued at 100 $/bbl would equal ˜ 1% of California’s total economic output. 21. CDC-DOGGR. 2008 Annual Report of the State Oil & Gas Supervisor. Technical report. California Department of Conservation, Division of Oil, Gas and Geothermal Resources: Sacramento, CA, USA, 2009.

20. ISO. ISO 14040: Environmental Management—Life Cycle Assessment—Principles and Framework; International Organization for Standardization: Geneva, Switzerland, 2006.

21. ISO. ISO 14044: Environmental Management—Life Cycle Assessment—Requirements and Guidelines; International Organization for Standardization: Geneva, Switzerland, 2006. 24. Herendeen, R.A. Net Energy Analysis: Concepts and Methods. In Encyclopedia of Energy, Cleveland, C.J., ed.; Elsevier: Amsterdam, The Netherlands, 2004; Volume 4, pp 283-289

22. CERI. Net Energy Analysis: An Energy Balance Study of Fossil Fuel Resources. Technical report. Colorado Energy Research Institute: Golden, CO, USA, 1976.

23. Pehnt, M. Dynamic Life Cycle Assessment (LCA) of renewable energy technologies. Renew. Energy 2006, 31, 55–71.

27. Levasseur, A.; Lesage, P.; Margni, M.; Descheacnes, L.; Samson, R. Considering Time in LCA: Dynamic LCA and its application to global warming impact assessments. Environ. Sci. Technol. 2010, 44, 3169–3174.

28. Mendivil, R.; Fischer, U.; Hirao, M.; Hungerbuhler, K. A new LCA methodology of technology evolution (TE-LCA) and is application to the production of ammonia (1950–2000). Int. J. Life Cycle Assess.2006, 11, 98–105.

29. Farrell, A.; Plevin, R.J.; Turner, B.T.; Jones, A.D.; O’Hare, M.; Kammen, D.M. Ethanol can contribute to energy and environmenal goals. Science 2006, 311, 506–508. 30. Cleveland, C.J. Energy quality and energy surplus in the extraction of fossil fuels in the US. Ecol. Econ. 1992, 1992, 139–162.

30. Gever, J. Beyond Oil: The Threat to Food and Fuel in the coming Decades, 3rd ed.; University Press of Colorado: Niwot, CO, USA, 1986.

31. Hall, C.A.S.; Cleveland, C.J.; Kaufmann, R. Energy and Resource Quality: The Ecology of the Economic Process; Wiley: New York, NY, USA, 1986.

32. Spitzley and Keolian outline more than 10 energy return ratios that have been used previously. These ratios differ in system boundaries included, in the quantity of interest, and in the inclusion or exclusion of various energy types (e.g., fossil energy return ratios that measure the leveraging of fossil energy streams in renewable energy systems). In many published studies, authors fail to specify exactly which energy return ratio is used.

33. Spitzley, D.V.; Keoleian, G. Life Cycle Environmental and Economic Assessment of Willow Biomass Electricity: A Comparison with other Renewable and Non- Renewable Sources. Technical Report CSS04-05R. University of Michigan: Ann Arbor, MI, USA, 2004.

34. Large amounts of data are available from the California Department of Conservation—Department of Oil, Gas, and Geothermal Resources (CDC-DOGGR). Databases contain well-level information from the late 1970s to the present, and field-level data are available in annual reports dating to 1915.

35. CDC-DOGGR. Summary of Operations and Annual Report of the State Oil & Gas Supervisor (Various). Technical report. California Department of Conservation, Division of Oil, Gas and Geothermal Resources: Sacramento, CA, USA, 1955–2005. 37. Caterpillar. Drilling rig repower. Brochure, Caterpillar Inc.: Peoria, IL, USA, 2003. http:// catoilandgas.cat.com/cda/files/823409/7/LEDW3153.pdf 38. Caterpillar. Caterpillar D397 Twelve Cylinder Specification Sheet. (Date illegible, approximately 1950). 39. EIA. Electric Power Annual 2007. Technical report. Energy Information Administration: Washington, DC, USA, 2009.

36. Schmidt, P.F. Fuel Oil Manual, 4th ed.; Industrial Press: New York, NY, USA, 1985. 41. Cheyenne Drilling Inc. Personal communication with staff of Cheyenne Drilling on energy consumption during drilling, 2006.

37. Cheyenne Drilling Inc. Drilling rig information, rig 1. Web page, Cheyenne Drilling, 2006.

38. Brandt, A.R. Converting oil shale to liquid fuels: Energy inputs and greenhouse gas emissions of the Shell in situ conversion process. Environ. Sci. Technol. 2008, 42, 7489–7495.

39. Azar, J.; Robello, G. Drilling Engineering; PennWell Publishers: Tulsa, OK, USA, 2007.

40. PSAC. 2001/2002 Well Cost Study. Technical report. Petroleum Services Association of Canada: Calgary, AB, Canada, 2000.

41. PSAC. 2002 Well Cost Study. Calgary, AB, Canada, 2001.

42. PSAC. 2005 Well Cost Study. Calgary, AB, Canada, 2005. Technical report. Petroleum Services Association of Canada: Technical report. Petroleum Services Association of Canada:

43. Gow, S. Roughnecks, Rock Bits and Rigs: The Evolution of Oil Well Drilling Technology in Alberta, 1883–1970; University of Calgary Press: Calgary, AB, Canada, 2005.

44. Smil, V. Prime Movers of Globalization: The History and Impact of Diesel Engines and Gas Turbines; The MIT Press: Cambridge, MA, USA, 2010.

45. Caterpillar Inc. Caterpillar D399 Marine Generator Set – Technical Specifications, 1980. https://marine.cat.com/cda/files/1014636/7/Spec+Sheet+-+Cat+D399+Genset.pdf

46. Horvath, A. Construction materials and the environment. Annu. Rev. Environ. Resour. 2004, 29, 181–204.

47. Brennan, J.R.; Vance, W.M., Section 9.17 – Oil Wells. In Pump Handbook, 3rd ed.; Karassik, I.J., Messina, J.P., Cooper, P., Heald, C.C., Eds.; McGraw Hill: New York, NY, USA, 2001.

48. Lyons, W.C.; Plisga, G.J. Standard Handbook of Petroleum and Natural Gas Engineering, 2nd ed.; Gulf Professional, Imprint of Elsevier: Burlington, MA, USA, 2005.

49. Takacs, G. Sucker-Rod Pumping Manual; PennWell Books: Tulsa, OK, USA, 2003. 55. Cited values of?t ranges from 85% to 93%, while?l ranges from 38% to 94% [54]. The lowest values were due to rubbing of the pump string against well casing in deviated wells [54]. I assume ranges from 80-90%.

50. Ayres, R.U.; Ayres, L.W.; Pokrovsky, V. On the efficiency of US electricity usage since 1900. Energy 2005, 30, 1092–1145.

51. Green, D.W.; Willhite, G.P. Enhanced Oil Recovery; Henry L. Doherty Memorial Fund of AIME, Society of Petroleum Engineers: Richardson, TX, USA, 1998.

52. Burger, J.; Sourieau, P.; Combarnous, M. Thermal Methods of Oil Recovery, 3rd ed.; Editions Technip, Gulf Publishing Company: Houston, TX, USA, 1985.

53. Brandt, A.R.; Unnasch, S. Energy intensity and greenhouse gas emissions from California thermal enhanced oil recovery. Energy & Fuels 2010, 24, 4581–4589.

54. CDC-DOGGR. Annual Report of the State Oil & Gas Supervisor. Technical report. California Department of Conservation, Division of Oil, Gas and Geothermal Resources: Sacramento, CA, USA, 2005.

55. Keesom, W.; Unnasch, S.; Moretta, J. Life Cycle Assessment Comparison of North American and Imported Crudes. Technical report. Jacobs Consultancy and Life Cycle Associates for Alberta Energy Resources Institute: Chicago, IL 2009.

56. Wang, M.Q. Estimation of Energy Efficiencies of US Petroleum Refineries (Plus Associated Spreadsheet). Technical report. Center for Transportation Research, Argonne National Laboratory: Argonne, IL, USA, 2008.

57. Brandt, A.R. Converting oil shale to liquid fuels with the Alberta taciuk processor: Energy inputs and greenhouse gas emissions. Energy Fuels 2009, 23, 6253–6258.