[ Excerpts from the Hall, Lambert, and Balogh EROI paper. You may want to read the original paper here since I’ve left out charts, figures, and text. In my opinion, EROI is important because it is due diligence – society ought to find out if there is any energy resource that can replace oil for transportation, since without transportation you can not build electricity-producing contraptions and you’re wasting rare earth minerals, steel, fossil fuels, and other finite materials making them. It is unlikely transportation can be electrified for reasons explained in my book When Trucks Stop Running: Energy and the future of transportation.
If decreasing numbers of trucks, rail, and ships will be running if oil can’t be replaced or heavy-duty vehicles electrified, our remaining energy should be used to clean up nuclear waste, superfund sites, the half million leaking mines, and other messes since future generations won’t have the energy to do so, lower our population ASAP to get within the carrying capacity of a non-fossil-fueled civilization (Plan B is bullets and disease), change our culture from one of consumption to one of sharing, teach different skills in schools to prepare the youngest generation, and prepare for going back to the age of wood (i.e. more insulation, gravity based water and sewage infrastructure that doesn’t require electric pumps where possible, and so on). Alice Friedemann www.energyskeptic.com ]
Charles A.S. Hall, Jessica G. Lambert, Stephen B. Balogh. 2014. EROI of different fuels and the implications for society. Energy Policy 64 (2014) 141–152
In the nations examined, the EROI for oil and gas has declined during recent decades. Lower EROI for oil may be masked by natural gas extracted/used in oil production. The EROI trend for US coal is ambiguous; the EROI for Chinese coal is declining.
All forms of economic production and exchange involve the use of energy directly and in the transformation of materials. Until recently, cheap and seemingly limitless fossil energy has allowed most of society to ignore the importance of contributions to the economic process from the biophysical world as well as the potential limits to growth.
This society must have an energy surplus for there to be division of labor, creation of specialists and the growth of cities, and substantially greater surplus for there to be wide-spread wealth, art, culture and other social amenities. Economic fluctuations tend to result, directly or indirectly, from variations in a society’s access to cheap and abundant energy
Today, fossil fuel re and economic expansion are eventually constrained by these higher prices (Jones et al., 2004). Economic growth and stability is dependent on not only the total quantity of energy accessible to society but also the cost of this energy to different sectors of that society.sources are among the most important global commodities and are essential for the production and distribution of the rest.
Economic production, exchange and growth requires work and consequently a steady and consistent flow of energy to do that work. Longer intervals of sustained economic growth in countries and the world have been punctuated by numerous oscillations; i.e. there are periods of economic expansion but also recession. In general, the growth of real GDP is highly correlated with rates of oil consumption (Murphy et al., 2011). Four out of the five recessions experienced since 1970 can be explained by examining oil price shocks (Hamilton, 2009; Hall and Groat, 2010; Jones et al., 2004). During periods of recession, oil prices tend to decline, eventually encouraging increased consumption. Alternatively, during periods of expansion, oil prices usually increase and higher energy consumption and economic expansion are eventually constrained by these higher prices (Jones et al., 2004).
Economic growth and stability is dependent on not only the total quantity of energy accessible to society but also the cost of this energy to different sectors of that society increases in the economic cost of energy (e.g. from five to ten percent) result in the diversion of funds from what is typically devoted to discretionary spending to energy acquisition (Hall and Klitgaard, 2012). Consequently, large changes in energy prices influence economies strongly.
Energy return on investment (EROI) is a means of measuring the quality of various fuels by calculating the ratio between the energy delivered by a particular fuel to society and the energy invested in the capture and delivery of this energy.
Much of the current EROI analysis literature tends to focus on the net or surplus for a given project, industry, nation, fuel, or resource, for example recent discussions on the “energy break even” point of EROI for corn based ethanol, i.e. whether the EROI is greater than 1:1. The apparently different results from this seemingly straightforward analysis generated some controversy about the utility of EROI. But, the variation in these findings is mostly the result of the choice of direct and indirect costs associated with energy production/extraction included within the EROI calculations: i.e. the boundaries of the denominator (Hall et al., 2011). The possible boundaries of the various net energy assessments evaluated in this study are illustrated in Fig. 1.
These and other boundary issues are addressed in Murphy et al.’s recent paper, Order from Chaos: A Preliminary Protocol for Determining the EROI of Fuels (Murphy et al., 2011). We clarify further the boundaries used in the EROI calculations given here into the following categories derived
Our research and that of Dale (2010) summarizes EROI estimates for the thermal energy delivered from various fossil fuels and also the electric power generated using fossil fuel and various other energy technologies. These initial estimates of general values for contemporary EROI provide us with a beginning on which we and others can build as additional and better data become available. We have fairly good confidence in the numbers represented here, in part because various studies tend to give broadly similar results.
Values from different regions and different times for the same fuels, however, can give quite different results. Given this, we present these values with considerable humility because there are no government-sponsored programs or much financial support to derive such numbers.
EROI values for our most important fuels, liquid and gaseous petroleum, tend to be relatively high. World oil and gas has a mean EROI of about 20:1
The EROI for the production of oil and gas globally by publicly traded companies has declined from 30:1 in 1995 to about 18:1 in 2006 (Gagnon et al., 2009).
The EROI for discovering oil and gas in the US has decreased from more than 1,000:1 in 1919 to 5:1 in the 2010s, and for production from about 25:1 in the 1970s to approximately 10:1 in 2007 (Guilford et al., 2011).
Alternatives to traditional fossil fuels such as tar sands and oil shale (Lambert et al., 2012) deliver a lower EROI, having a mean EROI of 4:1 (n of 4 from 4 publications) and 7:1 (n of 15 from 15 publication) (Fig. 2).
It is difficult to establish EROI values for natural gas alone as data on natural gas are usually aggregated in oil and gas statistics.
The other important fossil fuel, coal, has a relatively high EROI value in some countries (U.S. and presumably Australia) and shows no clear trend over time. Coal internationally has a mean EROI of about 46:1 (n of 72 from 17 publications) (see Lambert et al., 2012 for references). Cleveland et al. (2000) examined the EROI values for coal production in the United States. They found a general decline from an approximately 80:1 EROI value during the mid 1950s to 30:1 by the middle of the 1980s. Coal, however, regained its former high EROI value of roughly 80:1 by 1990. This pattern may reflect an increase in less costly surface mining. The energy content of coal has been decreasing even though the total tonnage has continued to increase (Hall and Klitgaard, 2012). This is true for the US where the energy content (quality) of coal has decreased while the quantity of coal mined has continued to increase.
The maximum energy from US coal seems to have occurred in 1998 (Hall et al., 2009; Murphy and Hall, 2010).
Meta-analysis of EROI values for nuclear energy suggests a mean EROI of about 14:1 (n of 33 from 15 publications)(see Lambert et al., 2013 for references). Newer analyses need to be made as these values may not adequately reflect current technology or ore grades. Whether to correct the output for its relatively high quality is an unresolved issue and a quality correction for electricity appears to contribute to the relatively high value given here.
Hydroelectric power generation systems have the highest mean EROI value, 84:1 (n of 17 from 12 publications), of electric power generation systems (see Lambert et al., 2012 for references). The EROI of hydropower is extremely variable although the best sites in the developed world were developed long ago (Hall et al., 1986).
We calculate the mean EROI value for ethanol from various biomass sources using data from 31 separate publications covering a full range of plant-based ethanol production e.g. EROI of 0.64:1 Pimentel and Patzek, 2005 for ethanol produced from cellulose from wood to EROI of 48:1 for ethanol from molasses in India (Von Blottnitz and Curran, 2007)). These values result in a mean EROI value of roughly 5:1 with an n of 74 from 31 publications (Fig.
Diesel from biomass is also quite low (2:1 with an n of 28 from 16 publications) (see Lambert et al., 2012 for references). The average is skewed in a positive direction by a handful of outliers (four EROI figures are above 30:1)
Wind power has a high EROI value, with the mean perhaps as high as 18:1 (as derived in an existing meta-analysis by Kubiszewski et al., 2010) or even 20:1 (n of 26 from 18 publications) (see Lambert et al., 2012 for references). The value in practice may be less due to the need for backup facilities.
We believe that outside certain conditions in the tropics most ethanol EROI values are at or below the 3:1 minimum extended EROI value required for a fuel to be minimally useful to society.
It should be noted that several recent studies that have broader boundaries give EROI values of 2 to 3:1 (Prieto and Hall, 2012; Palmer, 2013; Weissbach et al., 2013), although these are not weighted for the higher quality of the electricity when compared with thermal energy input.
A positive aspect of most renewable energies is that the output of these fuels is high quality electricity. A potential drawback is that the output is far less reliable and predictable. EROI values for PV and other renewable alternatives are generally computed without converting the electricity generated into its “primary energy- equivalent” (Kubiszewski et al., 2009) but also without including any of the considerable cost associated with the required energy back-ups or storage.
Poisson and Hall found that the EROI of conventional oil and gas has decreased since the mid-1990s from roughly 20:1 to 12:1, a 40% decline. The EROI of conventional combined oil-gas- tar sands has also decreased during this same period from 14:1 to 7.5:1, a decline of 46% (Poisson and Hall). Poisson and Hall’s estimated EROI values for Canadian oil and gas are about half those calculated by Freise and their rate of decline is somewhat less rapid (Freise, 2011; Poisson and Hall).
Poisson and Hall ‘s estimate of the EROI of tar sands is relatively low, around 4.5 (a conservative (i.e. high) estimate, using only the front end of the life-cycle); incorporating tar sands into total oil and gas estimates decreases the EROI of the oil and gas extraction industry as a whole (Poisson and Hall). These estimates would be lower if more elements of the full life-cycle (e.g. environmental impact) were included in the calculation.
Are studies at the regional level comparable to those at the national level and how do these “size up” when presented next to “international” studies that include a small subset of representative countries?
Energy analysts are not in agreement on what indirect costs should and should not be included in an EROI assessment. When complete systems are analyzed for solar PV installations, their financing, their operations and maintenance costs and their backups are included the energy costs are about three times larger than for just the modules and inverters. One very contentious indirect cost is the inclusion or exclusion of the energy cost of supporting human labor.
New work on the EROI for oil and gas produced by horizontal drilling and rock fracturing indicate that the EROI can be very high, in part because it is not necessary to pressurize the fields (e.g. Aucott and Mellilo 2013; Moeller and Murphy personal communication; Waggoner personal communication) but that these high values are likely to decline substantially as production is moved off the “sweet spots”.
The EROI for coal production in the US declined from 80:1 in the 1950s to 30:1 in the 1970s (Cleveland et al., 1984). During this time period, coal was mined almost exclusively in the Appalachian mountain region areas of the US using a combination of room and pillar mines with conventional and continuous mining methods. The coal initially extracted from these locations was a combination of anthracite and high quality bituminous coal, coal with high BTUs/ton. As the best coal was used first, the EROI for coal decreased over time.
The EROI of US coal returned to 80:1 by about 1990. This pattern reflects a shift in the quality of coal extracted, the technology employed in the extraction process and especially the shift from underground to surface mining. A shift in mining location, from Appalachia to the central and northern interior states of Montana and Wyoming and extraction method, from underground to surface mining (area, contour, auger, and mountain top mining techniques) have resulted in less energy required to mine and beneficiate coal. The energy content of the coal extracted, however, has decreased. The coal currently mined is lower-quality bituminous and sub-bituminous coal with much lower BTUs/ton (Hall et al., 1986; Hall and Klitgaard, 2012) The increased efficiency of surface mining seems to just about compensate for the decline in the quality of the coal mined.
Discretionary spending decreased with the energy price increases from 2007 to the summer of 2008. Oil prices hit an all time high of $147 per barrel in the summer of 2008 (Read, 2008). This extra 5-10% “tax” from increased energy prices was added to the US (and other) economy as it had been in the 1970s, and much discretionary spending disappeared (Hall et al., 2008). Speculation in real estate (in the US) was no longer desirable or possible as consumers tightened their belts because of higher energy costs (Hall and Klitgaard, 2012). The stock market crashed in September 2008 reducing market value by $1.2 trillion and forcing the Dow to suffer its “biggest single-day point loss ever” (Twin, 2008), and most Western economies have essentially stopped growing since.
Much of the discussion about “peak coal” (e.g. Patzek and Croft, 2010) involves changing mining technology and capacity, rather than the quantity and quality of coal that remains available for extraction. Peak coal will likely have the greatest impact on the world’s largest coal user, China. Nations with abundant untapped coal resources (i.e. the US, Australia and Russia) are likely to be less affected. The total recoverable coal estimated for the US alone is approximately 500 billion tons.
US coal production in 2009 was about one billion tons. Although it is difficult to predict future production technology, environmental issues, consumption patterns and changes in EROI, it appears that coal may be abundantly available through the next century.
Renewable energy sources: are not sufficiently “energy dense”, tend to be intermittent, lack transportability, most have relatively low EROI values (especially when corrections are made for intermittency), and currently, lack the infrastructure that is required to meet current societal demands.
If we were to replace traditional nonrenewable energy with renewables, which seems desirable to us in the long run, it would require the use of energy-intensive technology for their construction and maintenance. Thus it would appear that a shift from nonrenewable to renewable energy sources would result in declines in both the quantity and EROI values of the principle energies used for economic activity.
Although wind, apparently relatively favorable from an EROI perspective and photovoltaic (PV) energy, are currently the world’s fastest growing renewable energy sources, they continue to account for less than one percent of the global energy portfolio (REN21, 2012). Nevertheless there are many informal reports of PV reaching “price parity” with fossil fuels and to many the future of PV is very bright.
Proponents of EROI assessments using actual operational installations (rather than laboratory estimates) believe that, in order to portray renewable energy technology accurately, it is necessary to make note of the fact that these technologies are dependent upon (i.e. constructed and maintained using and therefore subsidized by) high EROI fossil fuels. Higher EROI values found in conceptual studies often result from assumptions of more favorable conditions (within simulations) than those actually experienced in real life. For example, English wind turbines were found to operate considerably fewer hours per month than anticipated (Jefferson, 2012).
Kubiszewski et al. (2010) infer that variations in EROI values, in the case of reported EROI values for wind energy, (between process and input output analyses) stems from a greater degree of subjective system boundary decision-making by the process analyst, resulting in the exclusion of certain indirect costs.
Also of concern is that wind and PV technology are not “base load technologies”, meaning that future large scale deployment, beyond 20 percent of the grid capacity, will likely require the construction of large, energy intensive storage infrastructures which, if included within EROI assessments, would likely reduce EROI values considerably. In the case of wind, the cost for inclusion within a wind EROI analysis requires not only the initial capital costs per unit output but also the backup systems required for the 70 or so percent of the time when insufficient wind is blowing. Thus, the input for an EROI analysis of wind and PV technology is by and large “upfront” capital costs. This is in sharp contrast to the less well known “return” over the lifespan of the system. Therefore, a variable referred to as “energy payback time” is often employed when calculating the EROI values of wind and other renewable energy sources. This is the time required for the renewable energy system to generate the same amount of energy that went into the creation, maintenance, and disposal of the system.
The boundaries utilized to define the energy payback time are incorporated into most renewable EROI calculations. Other factors influencing wind and PV EROI values include energy storage, grid connection dynamics and variations in construction and maintenance costs associated with the installation location. For example, off-shore turbines, while located in wet salty areas with more reliable energy-generating winds, require replacement more often. Turbines located in remote mountainous areas require long distance grid connections that result in energy loss and reduced usable energy values (Kubiszewski et al., 2010).
Policy implications
In conclusion, the EROI for the world s most important fuels, oil and gas, has declined over the past one to two decades for all nations examined. It remains possible that the relatively high EROI values for the natural gas extracted during, and often used for, the production of oil may mask a much steeper decline in the EROI of oil alone. Declining EROI is probably already having a large impact on the world economy (Murphy and Hall, 2010; Tverberg, 2012). As oil and gas provide roughly 60-65% of the world’s energy, this will likely have enormous economic consequences for many national economies. Coal, although abundant, is very unevenly distributed, has large environmental impacts and has an EROI that depends greatly on the region mined.
The decline in EROI among major fossil fuels suggests that in the race between technological advances and depletion, depletion is winning. Past attempts to rectify falling oil production i.e. the rapid increase of drilling after the 1970 peak in oil production and subsequent oil crises in the US only exacerbated the problem by lowering the net energy delivered from US oil production (Hall and Cleveland, 1981). Increasing prices, thought by most economists to negate depletion through increasing incentives for exploitation, cannot work as EROI approaches 1:1, and even now has made oil too expensive to support the high economic growth it once did. It would be tempting
from a net energy perspective, to recommend that we replace fossil fuels with renewable energy technologies as the EROI for fossil fuel falls to a level where these technologies become competitive. While EROI analyses generate numerical assessments using quantitative data that include many production factors, they do not include other important data such as climate change, air quality, health benefits, and other environmental qualities that are considered “externalities” to these analyses.
The energy intensive carbon capture and sequestration (CCS) required to reduce fossil fuel emissions to levels equivalent with that of wind or PV electricity production would reduce the final coal EROI value considerably ((e.g. Akai et al. 1997 in Dale, 2010 and Lund and Biswas, 2008). EROI figures do not take into account the high life-cycle greenhouse gas emissions from thermal electricity production, and coal fired systems in particular (Raugei et al., 2012). This could, with difficulty, be worked into future, more comprehensive EROI calculations.
Most alternative renewable energy sources appear, at this time, to have considerably lower EROI values than any of the non-renewable fossil fuels. Wind and photovoltaic energy are touted as having substantial environmental benefits. These benefits, however, may have lower returns and larger initial carbon footprints than originally suggested (e.g. the externalities associated with the mining of neodymium and its subsequent use in wind turbine construction). The energy costs pertaining to intermittency and factors such as the oil, natural gas and coal employed in the creation, transport and implementation of wind turbines and PV panels may not be adequately represented in some cost-benefit analyses. On the positive side, the fact that wind and PV produce high quality electricity needs to be considered as well.
Thus society seems to be caught in a dilemma unlike anything experienced in the last few centuries. During that time most problems (such as needs for more agricultural output, worker pay, transport, pensions, schools and social services) were solved by throwing more technology investments and energy at the problem. In many senses this approach worked, for many of these problems were resolved or at least ameliorated, although at each step populations grew so that more potential issues had to be served. In a general sense all of this was possible only because there was an abundance of cheap (i.e. high EROI) high quality energy, mostly oil, gas or electricity. We believe that the future is likely to be very different, for while there remains considerable energy in the ground it is unlikely to be exploitable cheaply, or eventually at all, because of its decreasing EROI.
Alternatives such as photovoltaics and wind turbines are unlikely to be nearly as cheap energetically or economically as past oil and gas when backup costs are considered.
In addition there are increasing costs everywhere pertaining to potential climate changes and other pollutants. Any transition to solar energies would require massive investments of fossil fuels.
Despite many claims to the contrary- from oil and gas advocates on the one hand and solar advocates on the other-we see no easy solution to these issues when EROI is considered.
If any resolution to these problems is possible it is probable that it would have to come at least as much from an adjustment of society’s aspirations for increased material affluence and an increase in willingness to share as from technology.
Unfortunately recent political events do not leave us with great optimism that such changes in societal values will be forthcoming.