“No Miracles Needed” by Mark Z Jacobson

The famous US environmentalist, Bill McKibben, starts his foreword to this book with the following quote

“This is among the most important books you’ll ever read, because it lays out in clear and frank terms the great problem of our age and the great solution”

That is quite an assertion! So it was with interest that I approached reviewing Mark Jacobson 's 350+ page book setting out how we could transition our energy system away from a dependency on fossil fuels to one entirely powered by Wind, Water and Solar (WWS).

Let me jump straight to the conclusion – this is most definitely a must read although, as I shall explain, there are other ways to access the important information it contains, which may better suit the already well-informed reader or practitioner.

The key conclusions are:

1. Because so much of our existing energy system uses highly inefficient fossil fuels the replacement by WWS technologies needs much less power. Globally average demand would be 20.4 TW of power in a fossil fuel “business as usual” 2050 scenario, but only 8.9 TW in the WWS 2050 scenario. That is 56% less energy demand. In other words, the WWS energy system will be only 44% of the size of the fossil fuel system.
2. 38% of the decrease is due to the inherent inefficiency of fossil fuels. A further 11% reduction in global energy demand arises from the very large energy consumption of the fossil fuels extraction, processing and transportation systems which will no longer be needed in a WWS energy system. We get a 49% reduction in end use energy demand by switching away from fossil fuels. This is sometimes referred to as “the primary energy fallacy” which is why fossil fuels proponents quote energy in fossil fuel equivalents to make the transition appear more difficult than it really is.
3. The model assumed an additional 7% overall improvement in efficiency from ancillary action like insulation. That is very conservative in my view and there is even further room for improvement.
4. In the WWS 2050 global system we will only have 12% losses from “shedding”, also known as curtailment. In effect we need to overbuild the system by this amount to ensure supply meets demand every 30 seconds. This is much lower than many folks who are anti-renewables quote.

I cannot state how important those headline results are in terms of guiding our decarbonization strategies and policies. In my review I will look at these points in greater detail and demonstrate why I believe that the analysis is correct and what it means for the UK.

It is important to note at the outset the parameters of the book. First of all, it is primarily concerned with changes needed to address energy-related climate impacts, so there is no discussion here of the 25% or so of climate change arising from our use of land. Second, it considers three aspects of the current and the WWS energy system: climate change effects, air pollution and energy insecurity.

In this review I will both be reviewing the book and walking through the methodology used to arrive at the key conclusion that a 100% renewables energy system is possible. The latter means that I will be going through some of the published papers that underpin the book, which I will reference this should you wish to see where the values come from.

To provide an assessment of the book it seems pretty self-evident that the central premise needs to be evaluated, so I make no apology for this. ?

Global warming

I always appreciate it when I learn something new, and Mark’s book gave me numerous new insights. One relates to the causes of global warming. Most folks are aware of the main greenhouse gases (GHGs) which we track – so-called Kyoto gases, since it was that Treaty that created an obligation to track them. These GHG’s are Carbon dioxide (CO2), Methane (CH4), Nitrous oxide (N2O) and several groups of industrial gases. Nitrogen trifluoride was recently added to the list.

Now I have been aware that there are other gases and aerosols that impact on climate change.? Water, for example, is a powerful greenhouse gas, but because it is largely in equilibrium in the atmosphere it is not monitored under the Kyoto Protocol. We then have aerosols which are not strictly speaking gases, but fine particles suspended in the atmosphere, which can have a warming or a cooling effect. Other GHGs may become more important in the future if we start to release these to the atmosphere, like Hydrogen (see a prior article of mine which touches on this subject). So, I was fascinated to see Mark, an atmospheric scientist, has quantified some of the effects of these other substances.

The Black Carbon is mainly due to the incomplete combustion of fossil fuels. It is the second larges cause of global warming after CO2. Brown Carbon is similarly produced from combustion of fossil fuels, but also biomass. Because these particles cause such a large global warming effect, and have short lifetimes, reducing their emissions is the fastest way to slow global warming.

Now we could be tempted to leave the “good” aerosols in the atmosphere since they have halved the effect of all the other warming gases. However these particles cause air pollution and health consequences, so the intention of the WWS model is to eliminate these as well.

Surprising omissions

Another effect of the focus on improving air quality is that bioenergy solutions that feature in many climate action proponent’s recommendations are eschewed here. Poor air quality kills 7 million people a year, and bioenergy doesn’t solve that problem – indeed it may make it worse. Another reasons for excluding biomass is that it isn’t carbon-neutral as lots of folks like to claim. Not only is there a long delay between the CO2 emissions when the fuel is burned, and when the replacement crop grows sufficiently to absorb a similar amount of CO2, but the watering, fertilisation, harvesting, processing and transporting of the biomass all require energy inputs, leading to emissions. Also, from a land-use perspective, dedicated biomass crops are very inefficient compared to the amount of energy that could be harvested by, say, solar panels or wind turbines.

Another energy source that is ruled out in the model is new hydroelectric plants. I can only assume that this is because of negative social and environmental impacts that new dams would cause, although I couldn’t find an explanation for this modelling choice. To be clear, some new pumped hydro and some existing uprating the output of existing hydro plants are considered, just not any major new dams.

Finally nuclear power, whether in the form of large power stations or small-modular reactors is also omitted from the model. Mark is quite clear on the reasons why:

"If the world’s all-purpose energy were converted to electricity and electrolytic hydrogen by 2050, the 9 trillion watts (TW) in resulting annual average end-use electric power demand would require about 12,500 850-megawatt nuclear reactors (31 times the number of active reactors today), or one installed every day for 34 years. Not only is this construction timeline impossible given the long planning-operation times of nuclear, but it would also result in all known reserves of uranium worldwide for once-through reactors running out in about 3 years. As such, there is no possibility the world will run solely on once-through nuclear energy by 2050."

Even if only 6.4 percent of the world's energy were supplied with nuclear, the number of active nuclear reactors worldwide would nearly double to around 800. Many more countries would possess nuclear reactors, increasing the risk that some of these countries would use the facilities to mask the development of nuclear weapons, as has occurred historically.

Mark then goes on to cite the catastrophic failure rate (1.5%) of nuclear plants, the dangers of proliferation of nuclear weapons and the risks posed by radioactive waste and the mining of Uranium as further reasons nuclear power is not desirable.? With mean energy costs of 16.6c kWh compared to 3.8c per kWh for onshore wind, cost also don’t favour nuclear. Worldwide, the time from planning to operation of most nuclear plants is 10-19 years, so the time needed to deploy is also a show-stopper, given the urgency to tackle climate change and the cumulative emissions reductions that alternative technologies could achieve.

WWS technologies

So, let’s look now at what is included in the Wind, Water, Solar technologies:

These are all the items of equipment that are modelled in the peer-reviewed studies that underpin this book. The most recent of which is the 145-country model which consists of an analysis of the feasibility of a 100% WWS energy system in each of the countries and regions, representing some 99.7% of all global fossil fuels emissions. [1] The fact that this book is based on numerous peer-reviewed papers adds considerable credibility to the proposals compared to the usual “how to fix the climate problem” nonfiction works.

Chapters 2-11 describe most of the technologies above in considerable depth, with numerous real-world examples. I must admit that I did find the 240 pages in this part of the book somewhat challenging because I already knew most of the details and because the headings, sub-headings, sub-sub-headings and sidebar “Transition highlights” sometimes added to a sense that the information was somewhat piecemeal.

That having been said, this is certainly a great and very up to date reference for specific technologies. The list of recent nuclear power plant projects and their outcomes would have been very helpful on the several occasions I have been in dialogue with fellow LinkedIn energy voices on that subject. The index is excellent – confirming that this is a useful reference.

One aspect of this technology walk-through is just how long many of the WWS technologies have been around. I learned for example that Scotland played a big role: sometime between 1832 and 1839 Robert Anderson built the world’s first electric vehicle; in 1887 Professor James Blyth built and operated the world’s first wind turbine in Glasgow and in 1888 James Readman invented an electric arc furnace to produce phosphorous in Edinburgh.

On electric cars I was intrigued to learn that, for the US:

“…by 1900, about one-third of all vehicles on the road were battery-electric. The public wanted electric vehicles because they were quiet, easy to drive, required few repairs, and did not produce pollution. Unfortunately, once Henry Ford started producing, in 1908, the Model T, which cost one-third as much as a battery-electric vehicle, the electric vehicle market collapsed. The expansion of gas stations and the greater range and top speed of gasoline vehicles through the 1920s sealed the fate of battery-electric vehicles for the next several decades.”

Energy Storage

I also found the discussions on thermal storage fascinating. The real-world project examples brought home the feasibility of this form of energy storage which I have personally had little experience of. You can imagine my surprise at the quantity of heat these systems can hold, their relatively low capital costs and their high rates of discharge.

There is an awful lot of scaremongering about the viability of 100% renewable energy systems due to a phenomenon called the dunkelflaute (German for “dark doldrums”) which is a period of time when the sun isn’t shining and wind isn’t blowing, which mean that wind and solar energy generation won’t be productive. Folks get particularly hot under the collar about the impact that would have on people’s ability to heat their homes, which is a very large proportion of domestic energy use in cold seasons.

The examples above, all based around district heating schemes in Denmark, show that very substantial heat stores can be created cost-effectively not only to tide over periods of low electricity availability, but also to reduce bills since the heat can be discharged when the electricity price is high. Some of these systems store a large proportion of the annual heat demand of the heat network:? 45%-50% at the Vojens pit store, 61% in Gram, another pit store. The heat for these systems is provided by solar collectors, so it is essentially “free”. These are long durations seasonal energy stores.

The principal energy storage mechanism will remain pumped hydropower storage:

“Excluding hydropower reservoirs, 97 percent of all electricity storage built on Earth prior to 2021 was pumped hydropower storage. Worldwide, about 530,000 potential pumped hydropower sites exist. These sites can store an estimated 22 million gigawatt-hours of energy for the electricity grid. This represents 100 times the energy needed to back up a 100 percent renewable electricity system worldwide.”

Pumped hydro has the benefit that it can be ramped up incredibly quickly, from no load to 100% in 15-30 seconds. That is much better than the fossil fuels world’s main dispatchable generation, gas fired combined cycle plants which take 10-20 minutes and coal or nuclear in 20 to 100 minutes.

Batteries form another very important storage option to match the demand from the grid in the short and medium term. There are a wide range of battery types but the book assumes the use of conventional lithium ion ones, although this is an area where there is considerable room for improvements in technology and cost, such as iron-air batteries. In terms of the battery design, it is assumed that each battery will hold in the order of 4 hours energy. So, assuming a maximum discharge rate of 10 kW, a battery could hold 40 kWh (not dissimilar to an electric car battery). The advantage modelling lots of small batteries is that they can work in parallel to attain much higher discharge rates, so if one wanted 1,000 kWh one could have 25 small batteries, with a maximum potential discharge rate of 250 kW. This compares to an alternative approach of one large 100-hour, 1,000 kWh battery which might have the same discharge rate of, say 10 kW. In the model one would simply use the number of 4-hour batteries in combination or in sequence to meet whatever the demand was at any time. Now 4-hour batteries exist today whereas larger 100-hour batteries do not.

The model on which the book is based assumes that the batteries are dedicated to the electricity storage task. However, there is one important technology called Vehicle-to-Grid which would effectively turn every car into a potential contributor of electricity when the grid demand was there.

“One study suggests that only 3.2 percent of vehicles in the United States are needed to smooth out US electricity demand when 50% of demand is supplied by wind” [2]

The key question

The discussions around technology are extremely informative and quite historical in their nature. Particularly entertaining was the story of the huge fight between Thomas Edison who favoured Direct Current electricity networks and Nikolai Tesla who saw the advantages of Alternating Current networks. However, I won’t go into the details here, suffice it to say that you will find these stories of skulduggery fascinating, and along the way you will learn about the benefits of AC if you weren’t aware of them before now.?

I want to turn my attention to what I consider the core of the book, which is the justification for the claim that a 100% renewable WWS energy system in 2050 is viable. The principal concern is that, for electricity, the supply at any moment matches the load – since that is the way all electricity systems have to work: supply and demand is balanced second by second. In other words that the system will be able to deal with a dunkelflaute period.

Now here I am going to do something that is strictly outside the purview of a book review. I am going to look at some of the original and more detailed analysis that Mark Jacobson and his team have made available online and drill down into a UK/Europe model. I am going to give a worked example to help explain the steps that are taken to arrive at a 100% WWS model. The description of this is in Chapters 12 and 13 of the book.

I am going to pick the UK for our example, but since the model depends on interconnections within larger regions to provide resilience, some of the data will be from the Europe region. You can access the specific paper with most of the numbers by checking out reference [3], which can be found in the first comment to this article.

Step 1 Work out the current and future energy use in BAU

The model that Mark and the team use does this for 7 fuel types, and 6 sectors expressed as an average power demand. This was 195 GW in 2018 for the UK and is forecast to rise, using the existing mix of generation to 232 GW. Ignore the last line of the table for the moment.

These figures are not the “nameplate” capacity of the various technologies, simply the total energy use in TWh divided by the number of hours (8,760) in a year to give a notional capacity so that comparison can be made across the WWS and BAU scenarios. In later steps we will be working with the actual “nameplate” capacity which is the maximum output that each technology type can achieve.

Step 2 Work out the demand reduction expected from electrification.

By way of example, consider cars. In petrol cars only 17 to 20% of the energy in the petrol is actually turned into useful forward motion (the tank to wheel efficiency). In a battery electric vehicle (BEV) the range of useful work done by the electricity in the battery is 64 to 89% (the plug to wheel efficiency). From these values we can derive a minimum and maximum fuel-to-electricity factor as in the range 0.19 (0.17/0.89) to 0.31 (0.2/0.64). By 2050 it is expected that the BEVs batteries will improve so that closer to the ‘best’ ratio 0.21 would be the average.? A similar process is used to calculate the fuel to electricity ratio of a hydrogen fuel cell vehicle, which is 0.52. ?We might also expect a small increase in the overall efficiency of cars of, say 4%, as their aerodynamics improve or their weight decreased, for example. This would be expressed as an improved efficiency factor of 0.96. See red box in Table 3 below.

Similarly, if we assume most heat goes to heat pumps in homes with a mean coefficient of performance, COP (useful energy delivered per unit energy in), of 4, with a range of 3.2 to 5.2, and that existing fossil fuel heating systems have a COP 0.8 we can see that the fuel-to-electricity ratio will be 0.2 (0.8/4). There is an assumption, modest in my opinion, that energy efficiency measures in residential oil use, like better insulation, controls and technology improvements will reduce demand by 16% giving us an improved efficiency factor of 0.84. See the blue box.

I’ve included the table with these factors below, omitting those for Agriculture/Fisheries and Military for brevity. The full table, S3, is on page 18 of [1].

Returning to Table 2, the total WWS 2050 UK load of 87.8 GW was arrived at by looking at each of the detailed BAU fuel demands for each fuel type and sector and multiplying these by the electricity-fuel factor and the improved efficiency factor.

The reduction in total power needed is remarkable in a WWS system. Instead of the forecast notional demand of 232.4 GW we only need 87.8 GW (see the last row of Table 2). That is 144 GW less than BAU, a reduction of 62% overall (k, below). Of that reduction, most of it, has come from the fuel-electricity ratio (h) and some from efficiency (i). A third source of reduction, in column (j), is the huge quantities of energy (11-13% of the global total) used to extract, process and transport fossil fuels which will not be needed in a WWS energy system (for example almost about 40% of all long-distance shipping is moving fossil fuels around).

These very substantial reductions in energy inputs to our system make the transition from fossils fuels much less challenging than one would think and save a huge amount of money.

Step 3 - Estimate the high-level WWS mix

Now that we know what the total demand in WWS terms looks like (88.7 GW) we can then work out which of the WWS generation technologies will provide that. The general rule of thumb adopted by Mark and the team is for around 90% of the demand to be met by solar and wind resources. Because wind resources are better in colder countries, there is a greater emphasis on this in the initial estimate of WWS resources needed at a regional level.

These estimates are checked to ensure that they only use a few percent of the land area available in the region and that the generation doesn’t exceed the total resource of that type for the region. The average costs among the generators, storage technologies and transmission/distribution costs should be kept low.

The first column is the BAU 2050 load reduced down to the 87.8 GW notional power to end users using the fuel and efficiency factors described in step 2. This has been allocated primarily to wind and then PV. Column (d) gives us the nameplate capacity of the equipment needed based on the transmission and distribution losses (b) and the capacity factors for the resource derived from the climate model (c). Given typical device capacities (e) we can then calculate the total number of devices (f).

Step 4 – calculating the energy available in 30 second intervals for 3 years

At this point, the analysis is done at a regional level, in our case, Europe. The reason for this is that the model assumes that long-distance grid interconnections will be an important tool to smooth out localised wind, tidal and solar generation. The UK is included in the Europe region as it has numerous interconnectors to Ireland, France, Norway, the Netherlands, and Belgium which accounted for 10% of electricity use in 2021 and had a capacity of over 7 GW [4]. It is possible that greater interconnection capacity will be needed in the WWS 2050 system.

This step involves calculating the amount of energy that the initial mix of WWS technologies would produce given actual weather simulation for the region being considered. ?Thus, if we have a certain amount of wind capacity the model works out the actual generation be from that capacity given the known windspeeds across the region in 30 second intervals.

This analysis is done for three years in order to smooth out variations that can happen year on year. This step used a prognostic global weather-climate-air pollution model (GATOR-GCMOM), which accounts for competition among wind turbines for available kinetic energy, to estimate wind and solar radiation fields and building heat and cold loads for each 30 second interval for three years (2050-2052) in each country.

This model ensures that there is a precise value for the energy available from the initial estimated technology mix. It takes into account all sorts of factors like competition for wind resources between turbines, the effect of roof mounted PV on the temperatures inside buildings. It also provides a “capacity factor” for each technology, which is the ratio between the theoretical output of a device, like a wind turbine, and the actual output given the windspeeds.

This model doesn’t just look at the energy supply, it also calculates how much heating or cooling energy is needed every 30 s to maintain the interior temperature among all buildings given the temperatures and solar irradiance etc., which is then used in the next step.

Step 5 – ensure that the energy available matches the end-use loads

This step of the model, LOADMATCH, takes the energy available in each 30 second interval, and the heating and cooling demands, and determines if it is sufficient to match each load (There is full description of this step in [1]- Note S4. Page 7).

Returning to Table 2, we had an initial 88.7 GW of demand allocated into 6 sectors. Each of those sectoral loads are now separated into (1) electricity and direct heat loads needed for low-temperature heating, (2) electric loads needed for cooling and refrigeration, (3) electricity loads needed to produce, compress, and store hydrogen for fuel cells used for transportation, and (4) all other electricity loads.

Each of these loads is then divided further into flexible and inflexible loads. Flexible loads include electricity and direct heat loads that can be used to fill cold and low-temperature heat storage (district heat storage or building water tank storage), electricity loads used to produce hydrogen (since all hydrogen can be stored), and remaining electricity and direct heat loads subject to demand response. Inflexible loads are all loads that are not flexible. ?

The LOADMATCH process follows the following rules: Loads subject to demand response can be shifted forward in time a maximum of eight hours. Loads subject to heat/cold storage can be met with storage or with electricity, either currently available or stored. Inflexible loads must be met immediately with electricity that is currently available or stored. Full details are in the reference, but the assumptions here seem sensible to me with 44% of the loads deemed inflexible.

The final figure in the table above is the amount of hydrogen estimated to be needed, which is 11 million tonnes for Europe. This is used for heavy transportation.

Step 6 Adjusted outputs from the LOADMATCH model

Once the model has been run completely through for the three years without pause, we know that there is sufficient energy to meet the loads every 30 seconds of the simulation. The results are shown in the table below:

One figure that jumped out of the page at me was the onshore wind potential for the UK, at 140%, appears to be higher than the maximum potential value. The reason for this is that the model determined that 1.4x times the initial onshore wind energy would be needed, the “Capacity Adjustment Factor”, to address the time sensitive loads However, that 1.4x figure was obtained at a Europe region, not at the UK, and must have been within the total Europe region onshore wind potential given that is one of the checks that LOADMATCH carries out.

While this looks like an error in the model, in practical terms this doesn’t affect the validity of the model in my mind because one could readily shift some of that wind offshore. The UK Committee on Climate Change suggest a potential for 29-96 of GW of onshore wind (see [6]-p 26) and this estimate falls in the middle of that.

Nevertheless, this reinforces the notion that, in the “real world”, one would refine the allocation of the WWS technologies based on all sorts of factors like supply chains, costs, public acceptance and so forth. Indeed, Mark and the team are at pains to point out that LOADMATCH is not an optimisation algorithm. It doesn’t work out what the cheapest or most rapid way to deploy combination of technologies is, which would require orders of magnitude greater computing power, but it simply determines that the energy derived from the weather model meets the loads in those 30-second windows. ??

Looking at the changes in the initial “minimal” allocation of WWS technologies in column (d) Table 5 and the final outputs in column S9 Table 7, we can see that we have increased the overall Nameplate capacity from 315 MW to 323 MW, an overall increase of just 2%. While there was a big increase in the onshore wind capacity, there was also a decline in the solar panels. There were also capacity additions of some solar and geothermal heat storage, shown in the bottom two rows of the table.

Now that we know the capacities of the final load balanced WWS system for the UK we can work out the energy supplied by each device. From those totals we can also calculate the percentage system losses. This is done at the regional level, Europe in our case.

We can see that the losses (d) from “shedding” aka curtailment are only 12% of the total energy produced in the system. The T&D and Maintenance losses (b) are 7% while the storage losses (c) are 3% in energy. The European and the UK figures are likely to be very similar, but the latter are not available.

Step 7 the benefits of a WWS system compared to BAU

Mark and his colleagues have also assessed the benefits that arise from the 100% renewables energy system.

Here are some of the headline results:

These are remarkable. The overall cost of the entire energy system has decreased by 68% (g)(h), a saving driven primarily by the cost of all those expensive but unused energy units in fossil fuel system, but also partly by the lower unit cost of energy in the WWS system (e).

Please note that the cost per unit of energy (e) is based on the capital cost of all the generation devices, additional storage, HVDC and shorter-distance transmission capacity, hydrogen production etc. ?(You can see these cost broken down in [3] Table 10). The total cost per tonne CO2 abated is $134 a tonne (481 MtCO2e/$65 bn a year). So we are not saying that the final consumer cost per kWh of energy will be 8.42 US cents – it will clearly be higher given the profit needed by the suppliers. However, with average wholesale prices 6.9 pKWh this month [ ] (8.7 US cents) the model may be suggesting a modest price rise.

It is assumed that air pollution mortality and morbidity will decrease by 90%, saving 12,500 lives a year in the UK. I won’t go into the “price per life” saved since that isn’t an area of expertise, but the paper does.

Comments on assumptions

The objective of the detailed walkthrough of the modelling step by step has been to establish the credibility of the “headline” behind the “No Miracles Needed”, claim that a 100% WWS system is feasible. Based on my analysis I am entirely comfortable that this has been a very detailed, considered and open piece of work from which one can conclude that a 100% WWS energy system is entirely viable and brings with it numerous benefits.

A UK WWS system would look something like:

While we have made some advances in terms of the renewables installations in the UK, it is clear that we need to accelerate the pace. In particular the model anticipates over 3.8 million rooftop PV installations (we currently have around 160,000) and with a current install rate of around 60,000 it would take 60 years to reach the target. An obvious next step would be to mandate that all UK new-build houses (not flats) have PV which would give an additional, say, 100,000 installations. On the wind side, the government is targeting 50 GW by 2030 [11], putting the UK well on the way to the target total and I think the objectives above are very do-able.

On the technical side, I think Mark and the team have done an excellent job in putting forward plausible, conservative estimates. I have already commented that I believe that considerably greater energy efficiency savings than the 8% modelled are realistic and achievable. On the offshore wind turbines, the assumed 5 WM size is very conservative. The UK government has estimated that the size of turbines commissioned in 2030, 2035 and 2035+ will be 15, 17 and 20 MW respectively. This is important because the turbines have much higher load factors (capacity factor net of availability) [12 page 14] which means they produce more electricity given certain wind speeds and the unit cost of generation is lower. Another figure that caught my eye was an assumption of a cost per kWh of $60.00 for lithium-ion batteries. While that was lower than the cost in the market at the time of publishing, there have been some recent announcements that prices will be as low as $57 kWh this year, so Mark’s prediction was, if anything, conservative.

?Where I do have a concern is the assumption that we could achieve 80% of the global WWS potential by 2030 and 100% by 2035 as set out in Chapter 14. I just don’t think that would be possible unless we go onto a “war footing”. Mark quite rightly points out that the biggest barrier to action are vested interests:

"Possibly the greatest challenge to overcome in transitioning the world to 100 percent clean, renewable energy is the challenge of repel. ling vested interests. Fossil-fuel companies have accumulated enormous. wealth and political influence since the start of the Industrial Revolution. As a result, they have been able to implement legislation that has given them an entrenched financial benefit in many countries through tax code breaks, direct grants, and subsidies. In addition, most of the existing energy infrastructure is a fossil-fuel infrastructure, and most people's day-to-day lives depend on that infrastructure (through transportation; building heating, cooling, and refrigeration; lighting, etc.). Further, over 25 million jobs worldwide depend on the fossil-fuel infrastructure."

This brings me to the notion of a just transition. This should not be seen simply a “sop” to workers to try to get them onboard and so help accelerate the transition but is instead a vital step to ensure that we don’t unwittingly cause huge damage to people and communities and we shift from one energy system to another. The closure of the UK coal industry in the 1990’s should give us a stark reminder of the impact – even today we have communities blighted by the loss of work and pride that came from being such an important part of the nation’s economy. For the UK Mark’s analysis is that 253,000 thousand jobs would be lost, but 477,000 jobs produced – the important thing is that the majority of those new jobs can and should employ folks from the communities affected by the losses.

One aspect of the book and the WWS analysis which I am uncertain about is the use of Hydrogen as a fuel in heavy transport (shipping and aviation) and as an energy store. I believe that the challenges posed by the physical properties of this gas make is a poor choice as an energy carrier. See my other review of “The Hype about Hydrogen” for more about that topic.

The paper [1]-S6 give a figure of 89.9 Tg of Hydrogen production per year (89 million tonnes). Globally we use 94 million tonnes in industrial processes [12], virtually all of it made from fossil fuels with large CO2 emissions (as well as supply-chain methane leaks). From an emissions-reduction perspective we would get a bigger reduction by using the electrolyser produced Hydrogen modelled to displace this industrial hydrogen than in transportation. We should note that we wouldn’t need 94 million tonnes of green hydrogen since around 42% is used in refining, which would largely disappear in a WWS world.

I don’t have a specific answer to this issue, but I would like to see the models incorporate 1) the production of hydrogen to displace grey hydrogen in a WWS future (e.g. around 50 million tonnes) and 2) an alternative to hydrogen in transport.

Chapter 14 sets out a useful list of policy actions that can help us get to 100% WWS

1. Build out the “smart” grid (2022)
2. No new fossil fuel plants (2022)
3. Electrify all new heating (including district heating) (2022)
4. Switch vessels to electricity, incentivise early retirement of fossil fuel ships (2022-2030)
5. Switch industrial heat to electricity (2023)
6. All new rail should be electric/hydrogen, new busses electric (2023)
7. All non-road vehicles should electrify (2023)
8. New truck transport electrify or hydrogen fuel cells (2023-2030)
9. New light duty vehicles (cars, vans etc) should electrify (2023)
10. New short haul aircraft, electrify (2023)
11. New long haul aircraft electrify or hydrogen fuel cells (2027-2035)

Policies will need to incentivise the early retirement of existing infrastructure to retire early to force the switch to WWS alternatives in time.

Some suggest that retiring fossil plants early will create stranded assets, incurring a large cost. However, averaged worldwide, the annual social cost of a WWS energy ?system is less than 10 percent that of a fossil system. As such, replacing a fossil plant before the end of its life saves substantial damage and money. Transitioning also increases the number of jobs available. Society thus benefits from stopping the operation of existing fossil-fuel plants and replacing them with new WWS plants as soon as possible.

Conclusion

This has been a very unusual book review. I have digressed to establish that the analysis is correct: we can deliver a 100% WWS world. For that reason, I highly recommend this book because this conclusion fundamentally influences our choices in how to address climate change.

I am less convinced by the timeline arguments. I do appreciate that we are fast exhausting out remaining Carbon Budgets, but the practical person in me simply doesn’t feel that a 2035 finish is credible, nor an 80% by 2030. 2050, on the other hand, may be much easier to achieve than we have assumed, because of the rapidly dropping costs of the WWS technologies and the growing political will. At the moment my instincts are that we will hit the early to mid-2040’s with the bulk of the transition completed and with some remaining gas plants or biomass plants “in reserve” to give us a bit of a safety margin while we prove to ourselves that WWS is as resilient as Mark’s excellent analysis would have us believe. To hit his proposed timeline would, I think “need a miracle”!

That having been said, I can’t emphasize enough how important the conclusion of this book are. I would thoroughly recommend fellow practitioners to explore the full array of resource that Mark and the Stanford team have made available [8] [9]. There’s lots there!

Mark is to be complemented for not only leading the studies that underpin this book but also for his efforts to engage with politicians and other influencers to get the 100% WWS message out there. One of those initiatives is The Solutions Project, which is well worth checking out [14].

Please note that all references [thus] are found in the first comment on this post.

Folks - if you have read this book please leave your own thoughts in the comments below - you may have picked out different aspects which others would find useful!

Do you agree with Mark that we shouldn’t have nuclear or biomass? Do you think the transition timeline is achievable? Please leave your comments!

You may also find my own textbook on energy and resource efficiency helpful - it's free to download :-)

Found this interesting/helpful? This is a link to all the book reviews so far with a brief summary and evaluation.