We offer the readers of ASH a translation of an article by Gail "Old Women" Tverberg (OurFiniteWorld), known for her systems approach, financial background and respect for physical economics. Good author, in short:-)
Why can renewable energy use models lie?
The energy needs of the world economy seem to be easy to model. Let's calculate the consumption: even in kilowatt-hours, even in barrels of oil equivalent, even in British thermal units, kilocalories or joules. Two types of energy are equivalent if they produce the same amount of useful work, isn't it?
For example, economist Randall Munroe explains the benefits of renewable energy in his video cover. According to his model, solar panels (if built to your liking) can provide enough electricity for yourself and half a dozen of your neighbors. Wind generators (also built to the level of absurdity, but of course) will provide energy to you and a dozen more neighbors.
However, there is a logical hole in this analysis. The energy produced by wind and solar panels is not exactly what the economy needs (at least for now). Wind and sun generate intermittent electricity, often available at the wrong time and in the wrong place. The world economy needs a variety of types of energy, these types must meet the engineering specifications of the most diverse systems of the modern world. Energy needs to be delivered to the right place and delivered to users at the right time of day or at the right time of the year. It may even be necessary to store the energy obtained from the sun and wind for several years (for example, you use a pumped storage power plant, and there is a drought in the region).
I think that the situation is similar to hypothetical scientists who decided, in order to increase the efficiency of the economy, to transfer 100% of the population from traditional food to grass and silage in 20 years. Cows, goats, sheep eat, don't they? Why can't people? Grass, no doubt, contains a ton of useful energy. Most types of grass appear to be non-toxic to humans - at least in small amounts. The grass seems to be growing pretty well. The grass can be stored for future use. Switching to the use of grass for food production seems to be worthwhile in terms of CO2 emissions. Unfortunately, grass and silage are not the kind of energy people usually consume. The fact that the great apes somehow did not evolve as herbivores is similar to the fact thatthat material production and transport in the modern economy are somehow ill-suited to intermittent energy from wind and sun.
Promotional video:
Putting grass in the human diet may well "work", but you need a different organism for that
If you look around, you can easily find herbivorous species. Animals with four-chambered stomachs thrive on a grass diet. These organisms often have continuously growing teeth because the silica in the grass tends to wear off the teeth. Perhaps, through genetic engineering, people can grow extra stomachs and add constantly renewed teeth. Other useful, but not very attractive adjustments to our body may be required, for example, to make the brain smaller (and the jaw larger). It takes too many calories to keep your brain active, you can't chew that much silage.
The problem with almost all current RES models is that the system is considered in a “narrow framework”. Only a small part of the problem is considered - usually only the falling price tags of panels and wind turbines (or “energy costs”) - and it is assumed that this is the only cost associated with changing the entire consumption pattern. In fact, economists have to admit that moving the economy to 100% renewable energy will require dramatic changes in society, similar to multi-chambered stomachs and ever-growing teeth, to switch to a 100% herbal diet. Your analysis needs a "wider scope".
If Randall Munroe took into account the indirect energy costs of the system, including the energy needed to rebuild existing power systems, his analysis would likely change. The ability of wind and solar energy to power both your own home and those of a dozen or so neighbors is likely to disappear. Too much energy will be used for the system to function as the equivalent of multi-chambered stomachs and ever-growing teeth. The world energy sector will work on renewable energy sources, but not in the same way as before. Roughly speaking, a smaller brain will think completely different thoughts.
Is “the energy used by a dozen of your neighbors” the right metric?
Before I go on about what went wrong with Munroe's model, I need to dwell briefly on his counting method. Munroe speaks of "the energy consumed by a household and a dozen neighbors." We often hear news about how many households a new power plant can serve or how many households were temporarily shut down due to the storm. The metric used by Munroe is very similar. But did he take everything into account?
In addition to households, the economy requires a variety of energy sources in many more places, including: in government for defense and law enforcement, in the construction of roads or schools, in farms for growing delicious food, and in factories for making useful things. It makes little sense to restrict the calculation to consumption in citizens' homes only. (In fact, Munroe is so streamlined in his calculations that it is not possible to figure out what exactly is included in his analysis. It seems that he counts only the energy that is in electrical outlets.) My independent analysis shows that directly in households only about a third of the total amount of all types of energy in the United States is consumed. The rest is consumed by private businesses and government bodies …
G. Tverberg's note:
My estimate of "about a third" is based on data from the EIA and BP. In terms of electricity, EIA data show that households in the United States use about 38% of total electricity generation. As for the fuel that is not used for transport and electricity generation, it is about 19%. Combining these two categories, we find that American households use about 31% of non-vehicle fuels. For transportation fuels, the best available data is BP's petroleum product statistics. According to BP, 26% of oil globally is burned as motor gasoline. In the United States, about 46%. Of course, some of this gasoline is not used for household needs: for example, police cars are usually gasoline, like small trucks used by businesses. Besides,The United States is a major importer of manufactured goods from China and other countries. The useful energy of fossil fuels embodied in these imports never appears in the US energy statistics.
One has only to adjust Munro's calculations to include the energy consumed by businesses and institutions, and we will have to immediately divide the specified dozen residential buildings into about three. Thus, instead of "energy sufficient for you and a dozen of your neighbors," you have to say: "energy for you and three or four neighbors." A dozen ("one order of magnitude" as the engineers would say) will evaporate somewhere. Moreover, the inclusion of social energy in the calculations is only the beginning of the path. As will be shown below, for a complete adjustment, you need to divide not by three, but by a much larger value.
What are the indirect costs from wind and solar renewables?
There are a number of indirect costs:
(1) The costs of delivering energy from renewable energy sources are much higher than those of other types of electricity, but in most studies they are either considered equal or averaged over the economy as a whole.
A 2014 study by the International Energy Agency (IEA) shows that the cost of transferring power from wind turbines is about three times the cost of power from coal or nuclear. As the share of wind and solar generating capacity in the total installed capacity increases, excess costs show an upward trend. Here are just a few of the reasons:
(a) The need to build more transmission lines, simply because the lines have to be designed to handle significantly higher peak loads. The power from the wind is usually available (see the link about the games with the capacity factor) from 25% to 35% of the time; the sun is available 10% to 25% of the time. {M. Ya.: According to BP, in 2018 the declared installed wind capacity was used by 25.7%, solar - by 13.7%. Miracles do not happen.}. Consequently, when these renewable energy sources operate at full load - for example, they store energy in a pumped storage power plant on a sunny and windy day - 3-4 times more transmission capacity of transmission lines is needed in comparison with continuously generating capacities.
(b) RES have, on average, a greater distance between the point of energy generation and the consumer. As an example, compare offshore wind turbines located 20-30 miles from the nearest community with a typical urban thermal power plant.
(c) Compared to fossil fuel capacity, the power generation of wind and solar power plants is much more difficult to predict - remember the proverbs about the incredible accuracy of modern weather forecasts. Consequently, the cost of energy dispatching increases.
(2) Due to the increase in the total length of the power transmission lines, the labor costs for maintaining these lines in a suitable and safe condition increase. This is especially unfortunate in arid and windy regions, where delays in maintenance of such lines can lead to fire.
In California, inadequate maintenance of power lines led to the bankruptcy of the PG&E power system. Consider how PG&E initiated two “preventative” blackouts, one of which affected about two million people. Texas power engineers report, "Our state's power lines have caused more than 4,000 fires in the past three and a half years." It's not just wind turbines. In Venezuela, wildfires along the 600-kilometer transmission line between the Guri hydropower plant and Caracas have triggered one massive blackout.
Of course, there are technical possibilities. The most reliable way is underground power lines. Even using insulated wire (hydroline) instead of bare wire can improve safety. However, any technical solution has its own price tag. These costs must be taken into account when modeling the development of renewable energy sources to the level of "the most desirable".
(3) Converting land transport to renewable energy will require huge investments in infrastructure. Of course, if only the uppermost layer of the “upper middle class” will use electric vehicles, then there is no problem. Understandably, the rich can afford both electric cars and (heated) garages / parking lots with dedicated electrical connections. It is clear that the rich will always find some way to charge their battery-powered car without a lot of hemorrhoids, and many of these amenities are already in stock.
The catch is that the less wealthy don't have the same opportunities. By the way, these “not the poorest” people are also very busy people, and they also cannot afford to spend hours waiting for the car to charge. This subset of consumers desperately needs inexpensive fast charging stations in many locations. The cost of fast charging infrastructure will likely need to include road maintenance taxes, as this is one of the costs that are now included in motor fuel prices in the US and many other countries.
{We are not even talking about the poor and the poorest strata of society. Their electric vehicle is, at best, a battery powered scooter. - M. Ya.}
(4) In conditions of a lack of reserve capacity, intermittent power supply increases the cost of material production. It is widely believed that intermittent generation can be relatively easily dealt with with simple organizational measures, such as “floating” daily / weekly / seasonal rates, “smart grids” with switching off household refrigerators and water heaters during peak loads, etc. These models are more or less justified if the system mainly consists of thermal power plants and nuclear power plants, and the share of renewable energy sources in generation is measured by the first percent.
The situation changes radically if the share of renewable energy sources exceeds these first percentages. We need chemical batteries that can smooth out daily peak loads, especially in the evening, when people come home from work and want to have dinner, and the sun - ah-trouble - has already set. The situation with wind turbines is even worse: there energy production can sink at any time, and not only because of the calm, but also because of the storm.
Batteries can help with daily cycle times and short-term outages, but renewables also have longer outages. For example, a severe storm with precipitation can simultaneously disrupt both solar and wind power for several days at any time of the year. Therefore, if the system is to operate only on renewable energy sources, it is desirable to have a reserve of energy for at least three days. In the short video below, Bill Gates talks pessimistically about the size of such a “battery” for a metropolis like Tokyo.
Even now, with a relatively low share of renewable energy sources in generation, we do not have devices capable of providing a full three-day backup. If the world economy switches exclusively to renewable energy sources, and electricity consumption per capita will still grow in comparison with the current one (electric cars, etc.), why do you think that it will become easier to create three-day uninterruptible power supplies?
But storing energy for three days is small compared to the seasonal cycle. Figure 1 shows the seasonal pattern of energy consumption in the United States.
Figure 1. US Energy Consumption by Month of the Year based on US Department of Energy data. “Rest” is total energy, minus electricity and transport energy. Includes: natural gas for heating, petroleum products for agriculture and all types of fossil fuels used in industrial production (petrochemicals, polymers, etc.)
Solar power production peaks in the United States in June, and lows from December to February. Hydroelectric power plants produce their greatest capacity during the spring flood, but their output varies from year to year. Wind energy changes unpredictably.
The modern economy cannot cope with power outages. For example, to smelt metals, the temperature must remain constantly high. Elevators shouldn't stop between floors, just because a storm has hit the wind farm. Refrigerators are required to cool so that fresh meat does not go bad.
There are two approaches that can be used to address seasonal energy problems:
(a) Rebuild industry so that in winter less energy is consumed for industrial production, and more is left for household needs. Smelt aluminum and burn cement only in summer!
(b) Build huge volumes of storage facilities, such as a pumped storage power plant, store energy for several months or even years.
Any of these approaches are extremely expensive. Approximately how to arrange a person on a second stomach using genetic engineering methods. As far as I know, these costs have not been included in any model to date {Gail is wrong. David McKay made such a model:
Figure 2 illustrates the high energy costs that can arise when adding a significant proportion of power redundancy. In this example, the "clean energy" that the system provides is essentially spent on maintaining the reserve in working order. The ERoEI parameter compares the useful energy output with the energy consumption.
Figure 2. The ERoEI chart of Graham Palmer, as reported by Australia Energy.
The example in Figure 2 is calculated for Melbourne, where the climate is relatively mild, and there is no hard frost or extreme heat. The example uses a combination of solar panels and cold standby chemical batteries in the form of diesel generators. Solar panels and chemical batteries provide 95% of the electricity in the system. Diesel generation is used during long-term interruptions and accidents and covers the remaining 5% of consumption. If the emergency diesel generators are removed from the model altogether, then more solar panels and more batteries will be needed. These additional batteries and panels will be used very rarely, but as a result, the ERoEI of the system will decrease even more.
Today, the main reason that the electricity system does not notice the costs of intermittent generation is the low share of wind and solar generation. According to BP, in 2018 the world generated 26614.8 TWh of electricity (398 watts of instantaneous power per capita). The contribution of wind was 1270.0 TWh (4.8%), the contribution of solar panels - 584.6 (2.2%). The total energy flow amounted to 13,864.4 million tonnes of oil equivalent (1,816 kg of oil equivalent per carcass per year), including 611.3 million toe from nuclear fuel. The share of wind in this huge volume is 287.4 million toe (2.1%), the share of solar electricity is 132.2 (1.0%). The wind and solar panels together gave for each earthling the equivalent of 1.5 car gas tanks: just under 56 kg of conditional oil.
The second reason why the electric power system does not yet notice the costs of renewable energy sources is that these additional costs are spread over the cost of the entire package of energy consumption, including for services of layered reservation with traditional sources of generation (coal, natural gas and nuclear power plants). The latter are forced to provide reserve capacities, including a “hot” reserve, without adequate cost compensation. This practice creates big problems for generating companies, and reserve capacities do not receive adequate funding. Traditional power engineers are forced to burn gas for free, without selling a single kilowatt-hour, only so that dim-green colleagues can sell wind and solar kilowatt-hours at a reasonable price and with acceptable overall power system reliability.
If, according to the ambitious plans of the Greens, the use of fossil fuels suddenly stops, all these reserve and base capacities, including nuclear power plants, will disappear. (The extraction of nuclear fuel, oddly enough, also depends on the fossil.) RES will suddenly have to figure out how to reserve capacity for their own money. This is when the problem of discontinuity becomes insurmountable. Strategic reserves of oil, oil products, coal, uranium can be stored for years, moreover with insignificant losses and relatively inexpensive; underground gas storage facilities are somewhat more expensive to operate; the costs of storing the generated electricity - whether in a pumped storage power plant or in chemical batteries - are incredibly huge. The latter include not only the cost of the system itself, but also the inevitable losses of electricity during pumping of pumped storage power plants and charging batteries.
In fact, the lack of financing of traditional capacities associated with the prerogative of RES for investment is already becoming an insurmountable problem in some places. Ohio recently decided to cut renewable energy funding and provide subsidies to nuclear power plants and coal-fired power plants.
(5) The cost of disposing of wind turbines, solar panels and chemical batteries is almost never reflected in the project budget.
It seems that in energy models there is a belief that at the end of their service life, windmills, panels and multi-ton batteries will dissolve by themselves in nature. Even if disposal costs are included in the estimates, it is often assumed that the cost of dismantling will be lower than the price of scrap metal. We are already discovering that the competent disposal of used waste is an expensive pleasure, and the energy consumption for recycling (especially metals and semiconductors) is often higher than all the energy sold to consumers during the operation of the installation.
(6) RES are not a direct replacement for many of the devices and processes that we actively use today. The list of things necessary for the exploitation of renewable energy sources is long, and much of this list is produced, at least for now, exclusively using fossil fuels. Helicopter wind turbine maintenance is a good example. Just don't try to convince us that heavy duty helicopters can also fly on batteries! Many of these processes or devices will not change for at least the next 20 years, which means that fossil fuels will be needed to keep renewable energy systems working.
In addition to servicing renewable energy sources, there are many other processes where there is no substitute for fossil fuel and is not visible in the future. Steel, fertilizer, cement, and plastic are four examples that Bill Gates mentions in his video. And we will also mention asphalt and most modern medicines. We'll have to change a lot and learn how to do without many of the usual goodies. It is impossible to construct either a road or a modern multi-storey building with the use of renewable energy sources, except perhaps with cobblestones. Probably, some of the materials can be replaced with wood, but will there be enough wood for everyone and will the world face the problem of massive deforestation?
(7) It is likely that the transition to renewable energy will take not 20 years as in the rosy forecasts of the Greens, but 50 years or more. During this time, wind and solar energy will act as a useful aid to the fossil fuel economy, but renewable energy cannot replace fossil fuels. This also increases costs.
In order for the production of fossil fuels to continue for the foreseeable future, resources and money will have to be spent at about the same rate as today. The delivery of fossil fuels still requires infrastructure: pipelines, refineries - and trained professionals. Miners, oil workers, gas workers, operators of thermal power plants and nuclear power plants, and many other workers of the "traditionally oriented" energy sector for some reason want to receive a salary all year round, and not only when there is a sudden snowfall, and solar panels are temporarily … Mining companies must pay off loans, received earlier for the construction of existing facilities. If natural gas is used as a winter reserve, new underground storage facilities will be needed. Even if the use of natural gas decreases, say, by a categorical 90%,then personnel and infrastructure costs - mostly fixed and little dependent on pumping volumes - will be reduced by a much lower percentage, say, by 30%.
One of the reasons why the transition to renewable energy will be long and painful is that in many cases there is not even a hint of how to get off the "oil needle". It is necessary to make changes in technology, and for this - to invent something new. Once invented, technical innovations need to be tested on real devices. When they tried, if everything is in order, it is necessary to build and establish technological lines for the mass production of new devices. It is likely that in the future it will be necessary to somehow compensate the owners of existing fossil-fueled devices and technologies for the loss of income or the cost of premature replacement of equipment. For example, forgive farmers for loans spent on the purchase of tractors and combines with internal combustion engines. If this is not done, the economy will collapse under the weight of bad debts. Only afterhow all these steps have been successfully implemented, we can talk about a real transition to a new technology. And so - for each specific technological chain!
These indirect costs make one wonder if there is any point in encouraging the widespread use of wind and sun in energy. Renewable energy sources can only reduce CO 2 emissions when they actually replace fossil fuels in electricity generation. And if renewable energy is just a politically correct add-on for a system that continues to devour fossil fuels, is it worth the effort?
Is the future of wind and solar power really better than the future of fossil fuels?
At the end of the video, Randall Munroe says that wind and solar power are infinitely available and fossil fuels are very limited.
In the last statement, I fully agree with Munro. Fossil fuels are very limited. This is because only natural energy carriers with a relatively low extraction cost are available to us.
The prices of finished products made with fossil fuels must remain low enough for the mass consumer to afford them. When we try to put into circulation resources with an increased extraction cost, mass demand shifts from discretionary goods (such as cars or smartphones) to everyday goods (such as food, heating, or clothing). The decline in demand for discretionary goods causes overstocking and a decrease in their production. Since cars and smartphones are manufactured using other goods, including fossil fuels, reduced demand for these goods leads to {MJ: hidden} deflation, including reduced energy demand (and prices). Therefore, the price of the resource balances on a patch of "already so expensive that few can afford" and "already so cheap,what you extract at a loss”, and everything is controlled by the presence (or rather the absence) of new energy deposits with an acceptable cost of extraction. It seems that since 2008 we have been in this state most of the time, experiencing a drop in real prices for oil and other resources.
{(M. Ya.: hidden deflation is masked by monetary emission, like "The economy is slowing down, let's throw Kuytsov as soon as possible!")}
Figure 3. Average weekly Brent price adjusted for inflation, based on EIA spot oil prices and US urban CPI.
Given this logic, it is difficult to understand why renewables should perform better or longer than fossil fuels. If the cost of RES without subsidies is higher than that of fossil fuels, RES will not develop. "It's already so expensive that few people can afford it." If renewable energy is subsidized, detaching from traditional energy, then traditional energy will cease to develop: "it is already so cheap that you extract at a loss." As shown above, RES in the foreseeable future cannot develop without the use of fossil fuels (for example, for the manufacture of spare parts for wind turbines or the construction / repair of power lines). Hence the conclusion: the development of renewable energy sources will inevitably begin to slow down, both with and without subsidies.
Do we believe in models too much?
The idea of using renewable energy sources sounds attractive, but the name is deceiving. Most renewable energy sources - with the exception of firewood, secondary biofuels (straw, cake) and dung-manure - are not renewable by themselves. In fact, renewables are highly dependent on fossil fuels.
{M. Ya.: the sun and the wind, they are, of course, practically eternal, but panels, batteries, turntables and even hydroelectric power plants / pumped storage power plants are by no means eternal. Twenty, thirty, well, a hundred years - PRESENT! We read from Kapitsa Sr.:
Interestingly, IPCC climate modelers and other climate change scarecrows seem to be fully convinced that the recoverable fossil fuel resources on Earth are, if not inexhaustible, very large. In fact, how much fossil fuels can actually be considered “recoverable” is one of the main problems of modeling, and this problem needs to be carefully studied. The volume of future production is likely to strongly depend on how stable the existing economic system is, including how stable the model of globalization of the world economy is. The collapse of the global system is likely to lead to a rapid decline in fossil fuel production.
In conclusion, I would like to emphasize that the social cost of renewable energy requires careful analysis. The hallmark of traditional energy (especially oil production) has always been huge profit margins. From these sky-high rates, through taxation, governments received enough funds to sponsor vital, but unprofitable sectors of the economy. This is one of the physical manifestations of ERoEI.
{M. Ya. ERoEI social versus standard ERoEI, read here:
If wind and solar energy really had such a high ERoEI, as some proponents counted, then these RES would not require subsidies: not only monetary, but also organizational, in the form of state preferences. In the meantime, as far as we know, the real ERoEI of RES is such that we are not even talking about taxing RES in favor of planned unprofitable sectors of the economy. Perhaps the researchers believe too much in their simplistic models.
Help about KIUM:
In the comments slipped that instead of the phrase "power available" (power input available) should be used abbreviation ICUF (Installed capacity utilization factor). Let us explain that the abbreviation KIUM CANNOT be used. There are at least three methods for calculating the "rated installed power" parameter for solar panels and wind turbines in the world:
Conditionally "Chinese". Does the panel at the back say "1kW" (maximum power)? Installed 1000 panels, which means the nominal installed power is 1 MW. You can even not connect to the network. Are the panels (on posts)? So they are "installed"! True, if you do not attach, then the ICUM will turn out to be 0, but the Chinese do not care about such trifles.
Conditionally "European Union". 1000 panels of 1 kW were connected according to the project to a 550 kW converter. This means the rated installed power is 0.55 MW. Above your head - sorry, the narrowest point of the system - you can't jump. This is the most correct counting technique, but it is not used everywhere. Well, the outlet power line should be 0.55 MW, despite the fact that on average per day the converter will give out about 0.22 MW in excellent sunny weather, and zero in snowfall.
Conditionally "USA". 1000 1kW panels in Northern California were connected to a 950kW converter. The average annual insolation coefficient for this particular location is 0.24. This means the nominal installed power is 0.24 MW. In a very successful year, if there is no snowfall, it is possible to generate 2.3 GWh, and ICUM = 108%!
