United States Deep Decarbonization Pathways Part I: Showing the Ways

United States Deep Decarbonization Pathways
Part I: Showing the Ways

A comment on a 2014 study from Lawrence-Berkeley and Pacific Northwest National Laboratories

Posted by Ed Leaver, 9 November 2015

Pathways to Deep Decarbonization in the United States 2050 Report, prepared by scientists and economists at Lawrence-Berkeley and Pacific Northwest National Laboratories, is part of the United States contribution to the international Deep Decarbonization Pathways Project, and has been submitted to the U.N. The report is available as a 100 page pdf, and is quite readable. I selectively quote from key sections, reproduce a few figures illustrating key results, and conclude with some brief commentary.

[Picture] [Picture] [Picture]

Big Wide Wonderful WyomingSharp End of the Hockey Stick Surface Coal Mine Near Gillette
For as far as the eye can see, Our Problem is Obvious

Figure 1: Good while it lasts.


“The Deep Decarbonization Pathways Project (DDPP) is a global initiative to explore how individual countries can reduce greenhouse gas (GHG) emissions to levels consistent with limiting the anthropogenic increase in global mean surface temperature to less than 2 degrees Celsius ( deg C)... Currently, the DDPP includes 15 research teams from countries representing more than 70% of global GHG emissions: Australia, Brazil, Canada, China, France, Germany, India, Indonesia, Japan, Mexico, Russia, South Africa, South Korea, the United Kingdom, and the United States. The research teams are independent and do not necessarily reflect the positions of their national governments.

“The initial results of this effort were published in September 2014 and officially presented as part of the Economic Case for Action session at the Climate Summit convened by UN Secretary General Ban Ki Moon in New York. That study, ‘Pathways to Deep Decarbonization: 2014 Report,’ included a chapter on deep decarbonization pathways in the U.S. The present report represents a continuation of the analysis in the DDPP Report, providing expanded results and greater detail on methods and data sources.”

Being a study undertaken jointly by two of our our National Laboratories and made part of an official presentation to the United Nations, one might surmise the report’s findings do, in fact, in some sense represent the positions and understanding of the United States federal government, or at least its executive branch.

Limiting the anthropogenic increase in global mean surface temperature to less than 2 degrees Celsius will require that global net GHG emissions approach zero by the second half of the 21st century. This Department of Energy Deep Decabonization Pathways study finds that it is technically feasible to reduce greenhouse gas emissions 80% below 1990 levels by 2050 in the United States, and that multiple alternative economic and technology pathways exist to achieve these reductions using commercial or near-commercial technologies. In addition to a business-as-usual baseline reference, four representative pathways were considered: “High Renewables”, (not particularly) “High-Nuclear”, “High-CCS”, and “Mixed”. Importantly, the study includes energy and emissions criteria for all aspects of the United States economy, not just the electricity sector most frequently considered.

It should be noted that while capital outlays for any of these scenarios is on the steep side of pricey – between $30 and $70 billion a year over thirty years – the net cost of some may in fact be negative, depending upon future prices of the fossil fuels (notably petroleum) being displaced.

Excerpts from the 2014 United States Department of Energy Deep Decarbonization Report

From the Abstract:

“Reductions are achieved through high levels of energy efficiency, decarbonization of electric generation, electrification of most end uses, and switching the remaining end uses to lower carbon fuels. The cost of achieving these reductions does not appear prohibitive, with an incremental cost to the energy system equivalent to less than 1% of gross domestic product (GDP) in the base case (“Mixed”). These incremental energy system costs did not include potential non-energy benefits, for example, avoided human and infrastructure costs of climate change and air pollution. The changes required to deeply decarbonize the economy over the next 35 years would constitute an ambitious transformation of the energy system. However, this study indicates that these changes would not necessarily entail major changes in lifestyle, since the low carbon pathways were designed to support the same level of energy services and economic growth as the reference case based on the U.S. Department of Energy’s Annual Energy Outlook. Starting now on the deep decarbonization path would allow infrastructure replacement to follow natural replacement rates, which reduces costs, eases demand on manufacturing, and allows gradual consumer adoption.”

2.3 Biomass Budget

Page 10. The study’s primary basis for biomass availability and costs is the DOE Billion Ton Study Update (BTS2), which includes resource potential estimates to 2030 for purpose-grown energy crops, agricultural and forest residues, and waste products, with some adjustments:

First, currently used resources in the Energy Information Administration’s Annual Energy Outlook 2013 (AEO) reference case were removed from the BTS2 estimates. These include fuel wood, mill residues, pulping liquors, and forest waste resources currently used primarily by industry in combined heat and power (CHP), power generation, and direct fuel applications. The PATHWAYS modeling continues to satisfy this current demand and does not make these biomass resources available for other applications in the future.

Second, the quantity of purpose-grown energy crops is constrained to a level (371 MMT, million metric tons) that does not result in indirect land use change GHG emissions... The composition of purpose-grown energy crops nationally is intentionally altered over time in the analysis, transitioning land currently used for corn ethanol production to second-generation energy crops (perennial grasses and woody purpose-grown feedstocks). With the remaining BTS2 biomass resources included, the upper limit on dry biomass supply in this report is 1,081 million metric tons, with a total primary energy value of 18.5 EJ.

Because biomass is a versatile energy feedstock that can displace different kinds of fossil fuel, the amount available with zero or low net lifecycle CO2 emissions, and its allocation to different forms of final energy supply, has a strong impact on other aspects of the energy system. In this study, biomass supply is used primarily for production of renewable gas and liquid fuels. Negative emissions bioenergy-CCS is applied only to bio-refining in the high CCS case.

3.3 Pathway Determinants

Primary energy for electric generation
Page 15. Electricity generation, including that used for production of intermediate energy carriers, becomes the dominant form of delivered energy in all deep decarbonization cases... the effects of generation mix are explored using “corner cases” with high renewables, high nuclear, and high CCS generation portfolios, plus a mixed case includes roughly equivalent generation from all three decarbonized options.
Electric load balancing resources
Page 15. The choice of primary energy for generation strongly affects electricity balancing requirements. For systems with high levels of inflexible generation (e.g., variable wind and solar, conventional baseload nuclear), a variety of balancing strategies are needed to maintain reliable system operation, including regional coordination, natural gas generation, curtailment, energy storage, and flexible loads. In this study, power-to-gas hydrogen and synthetic natural gas production are also used as balancing resources, providing low carbon fuels in the process... While these electric fuels may be inefficient from a primary energy perspective, their ability to operate flexibly reduces curtailment, which represents a system-wide inefficiency caused by large amounts of non-dispatchable generation. When this flexible load reduces curtailment, it can provide significant value as a component of an integrated energy system, despite its potentially high cost when viewed in isolation.
Fuel switching and efficiency
Page 15. Energy efficiency is widely considered the first option to pursue in a low carbon portfolio, with value independent of other pathway determinants. In deep decarbonization cases, coordinating end use choices with the other design choices (e.g., whether CCS exists, how biomass is allocated) is required to make optimal tradeoffs between fuel type and efficiency level from the standpoint of cost and emissions. In this study, significant efficiency improvements come from thermodynamic advantages inherent in certain kinds of fuel switching (e.g., from internal combustion to electric drive train vehicles, from natural gas heat to ground-source heat pumps).

3.4 Four Deep Decarbonization Scenarios


Figure 6: Pathways Determinants: Critical Elements that Determine the Features of a Low Carbon Energy System

Page 16. The four deep decarbonization cases created for this analysis represent a range of pathways that result from significantly different technology choices organized around the three primary energy choices for electricity – renewable energy (High Renewables Case), nuclear (High Nuclear Case), and fossil fuels with CCS (High CCS Case). The Mixed Case includes a balanced mix of all three primary energy resources. All cases have similar strategies for and levels of energy efficiency. The four cases are intended to illustrate a broad suite of consistent, interrelated technology choices, while still remaining tractable for purposes of presentation. They are not intended to be exhaustive.


Figure 7: Four representative deep decarbonization scenarios
Ed’s note:
In the far right Generation Mix column, the upper-left orange region of the Renewables, Nuclear, and CCS scenarios are likely intended to be colored yellow, representing solar fusion energy, rather than orange, which represents uranium fission. And at 45.4% wind+sun and 40.3% uranium, “High Nuclear” isn’t particularly high compared (for instance) with modern-day France, which today gets about 75% of her electric power from fission. This realistically high nuclear case was not considered in this (representative but non-exhaustive) report.

Table 7: Scenario Summary for 2014, 2050 reference case, and four 2050 deep decarbonization scenarios

  High   High   High
Indicator Units   2014  Reference Mixed   Renewables  Nuclear  CCS

Intensity Metrics
US population Million 323 438 438 438 438 438
Per capita energy use rate GJ/person 211 183 123 125 121 128
Per capita emissions t CO2/person 16.0 12.9 1.7 1.7 1.7 1.7
US GDP (billions 2012 dollars)  B 2012$ 16,378  40,032   40,032  40,032   40,032   40,032
Economic energy intensity MJ/$ 4.17 2.00 1.35 1.37 1.32 1.40
Economic emission intensity kG CO2/$ 0.31 0.14 0.02 0.02 0.02 0.02
Electric emission intensity gCO2/kwh 510.9 413.5 13.5 16.0 23.4 54.7

Energy Intensity Metrics for the four scenarios relative to 2014.
Ed’s Notes:
  1. MMT = 1 million metric tonnes. 1 metric tonne = 1000 kg = 2,200 lb.
  2. EJ (exajoule) = 1018 Joules = 278 TWh = 278,000 GWh
  3. From the Intensity Metrics, the four deep decarbonization scenarios all shoot for net US energy reduction of about 20% in 2050 relative to 2014, achieved mainly though increased energy efficiency and switching to more energy efficient fuels.
  4. From Electric Generation, despite the 20% decrease in total energy consumption, total net electricity production is required to double by 2050 in three of the four cases. The fourth, High CCS, has separate issues all its own.
  5. These four scenarios, while illustrative, are by no means exhaustive. Infinite possibilities.
  6. There are tradeoffs. For instance, relative to High Renewables, High Nuclear (being reliable) is able to avoid more expensive SNG storage, and also use somewhat more natural gas in electricity generation, which results in somewhat higher CO2 emissions in that sector. There is also marked difference in the ratios of electricity and H2+SNG used in ground transport, with High Renewables using substantially (46% vs 20%) more electricity directly in passenger electric vehicles in return for some smart-grid storage benefit. The High Nuclear scenario sees more of those vehicles being powered by hydrogen. Total net electricity generation and per capita emission are the same in each case.
  7. Per capita energy use with nuclear is slightly less than with renewables, possibly due to the relatively energy-inefficient synthetic methane (SNG) component of the High Renewables scenario. (Energetically, Carbon Capture and Storage is even worse.)

4.4 Cost

Incremental energy system costs – incremental capital costs plus net energy costs – exhibit a broad range in 2050, reflecting the significant uncertainty in technology costs and fossil fuel prices over such a long timeframe. Under base assumptions of technology costs and fossil fuel prices, the median value of incremental costs ranges from $160 billion (2012 $) to $650 billion across scenarios, with the difference driven primarily by the relative quantities and prices of residual natural gas and petroleum fuels remaining in the energy system in 2050.1 The average median value across cases is just over $300 billion.

Based on an uncertainty analysis of key cost parameters, the interquartile range of incremental energy system costs extends from negative $250 billion to $1 trillion... (Figure 12). To put these numbers in context, the (modeled) activity drivers that drive energy service demand in all of the cases are consistent with a U.S. GDP that grows by a real annual average rate of just over 2% per year over the next four decades, to around $40 trillion in 2050. The average 75th percentile estimate of net incremental energy system costs ($730 billion) across cases is equivalent to 1.8% of this GDP level. The average 25th percentile value is negative $90 billion.


Figure 12: Incremental Energy System Costs in 2050, error bars show the 25th and 75th percentile values.
Ed’s note:
Negative values for some of the 25th percentiles suggest we might actually come out ahead, although there will be no way of knowing in advance: the $0 level incorporates projected intrinsic costs of fossil fuels in the baseline bau case. As fossil fuel usage drops in the decarbonization scenarios, fossil fuel intrinsic costs will likely drop. It will be for economists to estimate by how much as we adjust market signals (carbon tax, cap-and-trade, fee-and-dividend) to obtain the desired decrease, if such are the mechanisms we wish to employ.

(End Report Excerpts)


Finally, we have here one study that compares different technology pathways using the same basic assumptions and models. As indicated under “Electric Generation” in Table 7, absent enormous carbon capture and storage, United States electric power generation will most likely need to double over the next 35 years. This new electric power must be essentially emissions-free; such clean generation must replace nearly all existing fossil generation at well. This by itself is a monumental undertaking, and is the focus of much of our discussion.

One interesting aspect of U.S. Deep Decarbonization Pathways report is it’s choice of energy storage which, while critical to smooth renewable intermittency, can be quite useful for nuclear load balancing as well. The study foregoes the usual high capacity storage systems – Pumped Hydro and Compressed Air – in favor of hydrogen (H2) and synthetic natural gas (SNG). Though not a particularly energy-efficient storage medium compared with CAES or pumped hydro, H2+SNG can similarly be generated during periods of over-supply of high-penetration intermittent wind and sun, and stored in existing natural gas distribution systems (with upgrades), allowing existing gas distribution infrastructure and peaking generators to transition from natural to synthetic gas with least additional capital investment.

Existing gas infrastructure cannot safely handle more than about 7% hydrogen, is why the interest in synthetic methane. The “High Nuclear” scenario anticipates using its additional hydrogen as motor fuel, transported largely in trucks to service stations just like today’s gasoline. (Only cryogenic...)

For those curious about nuclear technology, the report takes a conservative approach and apparently incorporates only the Generation III light-water reactor plants available today. These are modeled to run as steady baseload generators, with low-load power variations shunted to hydrogen generation for transport use.

For those who hold the cheapest energy is that you don’t use, the Deep Decabonization Pathways study does not disappoint. In Table 7 we see a healthy 43% drop in personal energy consumption with 36% population growth, for a net 20% decrease in overall US energy consumption. That is probably about as good as we’re going to get, politically, in the U.S.

The report does not suggest what kind of market signals we might adopt to effect that kind of savings. In 1859 William Jevons showed that increased utilization efficiency, absent an increase in price, can actually lead to an increase in consumption rate of the resource, rather than the desired saving. This is the now-classic Jevons Paradox; the dwindling resource Jevons considered was... British coal.2

Energy transition is an expensive undertaking. But we’re the richest nation on Earth: we can do it. Here. But carbon emissions and global warming are a global problem: see Figure 1. Solving our local problem will not do the U.S. a shred of good if the Rest of the World cannot solve their local problems as well.

And that is where I have a minor quibble with the U.S. Deep Decarbonization Pathways report. Gas backup or energy storage on the grid level is absolutely required for substantial intermittent renewable penetration, but there’s little reason to believe the proposed H2+SNG storage solution for the United States will scale to the Rest of the World. Don’t get me wrong: the US has extensive gas infrastructure and we might as well use it. The Rest of the World does not, and perhaps might be mistaken to acquire one: where natural gas distribution infrastructure is built, natural gas will be produced and burned. In many repects natural gas is a Good Thing. But there is such a thing as having too much, and at some point we should seriously consider just leaving our gas in the ground. The United States might carefully consider which of our own abatement technologies might be most economically adapted for export.

In that regard Figure 2 illustrates wind’s intermittent requirement for expensive gas backup (natural or synthetic), or for storage:


Figure 2: Wind noise on the Irish Grid, 2011
Upper: Demand (black) and estimated wind power generation (green). Mean demand is 2813MW. Mean wind was 467MW with capacity factor 30% and 17% penetration. The correlation between wind generation and demand is close to zero ( 0.05).
Lower: Effective demand (Demand - Wind) as seen by conventional plant. Effective demand is 38% more volatile than system demand.

Source: Joseph Wheatley Quantifying CO2 savings from wind power, Energy Policy Volume 63, December 2013, Pages 89 - 96 doi:10.1016/j.enpol.2013.07.123, with additional commentary at Quantifying CO2 savings from wind power redux: Ireland 2012, and more source material here.

Wheatley’s articles are fascinating. Ireland does have gas, so wind backup is not a problem. But Irish gas is more expensive than Irish peat or coal, so clean, lower-emission gas is what wind displaces when it blows. This is a problem, if one’s goal is to actually reduce carbon emissions, rather than to just deploy wind. But that’s what Renewable Portfolio Standards are for: to deploy renewable resources. Emission reductions are a very welcome knock-on, to the extent they occur.

Coal and peat and lignite have other problems when asked to ramp rapidly enough to balance wind, and even a few hours storage can be very beneficial. But wind+storage does add cost to otherwise finely operating thermal generation, and there is no readily apparent reason to expect the Rest of the World to foot the bill for such added expense when they’re strapped enough just trying to supply reliable generation in the first place.

So we also look for clean energy solutions the Rest of the World, not just our isolated corner, can afford. Here’s the problem: U.S. Deep Decarbonization Pathways illustrates a whopping 43% reduction in per capita U.S. energy consumption over 2014 levels. That will be doable, if not easy. But let’s put it in perspective: Germany has a 2014 per capita energy budget about one-half that of the United States, and Bulgaria’s is just one-third.

There are nearly 7.4 billion people in the world today. By 2050 there are projected to be over 9 billion before growth starts tapering off. Historically, there have been but two checks on human population: the Four Horsemen, and plentiful energy. Energy plentiful enough to provide healthy food, decent living accommodations with refrigeration, healthcare, and enough leisure time for a man, his wife, and his kids to educate themselves. Educate themselves, and accumulate enough wealth to support the man and his wife in their old age without needing a plethora of children to help.

In a very real sense energy – reliable, useful energy – is wealth. And personally, I’d prefer to see every person on this planet wealthy enough to limit our collective population by using that wealth in healthy, productive, and caringly civilized manner. As population control, The Four Horsemen are vastly overrated.

Yet today some 1.3 billion of us subsist on wood and dung and hand-collected coal burnt on indoor-fires, and have no access to electricity at all. That is poverty. Energy poverty. United States’ profligacy notwithstanding, what the Rest of The World needs is to consume more energy, not less.

Here’s the reality: if the U.S. were to reduce our per capita energy consumption by another 23% beyond that suggested by U.S. Deep Decarbonization Pathways, to reduce our 2014 consumption by two-thirds to match that of Bulgaria, and the Rest of the World were to raise their consumption to that level, net global energy consumption would more than double, from today’s 15.4 PWh/yr to 37 PWh/yr.
(1 PWh = 1012 kWh.) If instead our future 9.3 billion person world strives to achieve a living standard, measured by energy consumption, similar to that of today’s Germany, our global energy requirement will more than triple:


Figure 3: Going it alone: US vs The World.

The reality is that such living standards are precisely what the Rest of the World aspires to. Such aspirations are not unreasonable. Nor will they be denied: energy is plentiful, one needs but extract fuel from the ground and burn it.

Of course there are other sources. Hydropower is good, and reliable if you can get it. Like anything else hydro has environmental cost; there’s no free lunch. Neither is hydro available everywhere. Fossil fuels are.

So are wind and sun and uranium – although uranium is so incredibly energy dense and transportable, and refining, recycling, and eventually sequestering any final waste is such an undertaking, that very few of the 36 countries who use the stuff actually bother digging it up themselves.

But again, wind and sun require either extensive storage, or essentially 100% fossil and/or biomass backup. All of these cost, and if in combination they end up costing more than coal, coal it will be.

Figure 45 suggests that of the four scenarios considered, “High Nuclear” has the best shot at coming in cheaper than fossils, although such conclusion is by no mean certain and can hold only if one can take the long view about energy investment. It also can hold only if fossil fuel prices remain high, which is by no means assured if world-wide consumption drastically drops (as it must), and there are insufficient local penalties to discourage fossil fuel use. Ideally – and seemingly in practice as well – one would like clean reliable energy that is intrinsically cheaper than coal. And that’s a tall order.

But not necessarily out of reach. Nuclear power is robust and reliable and reactors can last a long long time. It’s actually a pretty good deal, if you can afford it. We’ll have more on this, and on how storage can contribute, as we continue this series.

Suggested Reading :

  1. Energetic Africa Breakthrough Institute Staff, 7 April 2014.

  2. Obama Thinks Solar Power Will Boost Kenya; Kenyans Aren’t So Sure NPR, July 23, 2015.

  3. Kenya’s Studied Approach to a Nuclear Future ANS Nuclear Cafe, Will Davis, April 8, 2014.
  4. South Africa to start nuclear procurement World Nuclear News 15 July 2015.
  5. Are Solar Microgrids a Step on the Ladder Towards Grid Access? Alex Trembath, Breakthrough Institute, 16 Dec 2014.

  6. Potential for worldwide displacement of fossil-fuel electricity by nuclear energy in three decades based on extrapolation of regional deployment data. Staffan Qvist and Barry Brook (2015) PLoS ONE 10(5): e0124074. doi: 10.1371/journal.pone.0124074

  7. Google Images for Wyoming Coal and Images for Appalachian Coal and Images for Australian Coal and Images for China Coal.

1Petroleum fuel prices are significantly more expensive than natural gas by 2050 in the AEO 2013 Reference Case. Thus, scenarios in which more petroleum fuels are displaced are lower net cost.

2With respect to Jevons’ finding that resource price must increase if increased efficiency is to effect an actual decrease in consumption, I personally favor the fee-and-dividend approach advocated by Citizens Climate Lobby for penalizing emissions with a stiff fee imposed on fossil fuel import and extraction, the fee then refunded to American households. But there certainly are other options.