Green advocates and climate change alarmists alike insist that the world shift to using only non-polluting, renewable energy sources, and the sooner the better. What is seldom mentioned is the enormous cost of retooling the world's energy infrastructure to use intermittent, unreliable wind and solar energy. A recent two part paper, appearing in Energy Policy, makes a reasonable attempt at stating the requirements to fix humanity's fossil fuel addiction and go all green. The analysis found that, to provide roughly 84% of the world's energy needs in 2030, would require around 4 million 5 MW wind turbines and 90,000 300 MW solar power plants, with the remaining 16% coming from solar photovoltaic rooftop systems, geothermal, tidal, wave and hydroelectric sources. Some quick back-of-the-envelope calculations show why the world economy cannot afford to go totally green.

Mark Z. Jacobson and Mark A. Delucchi have provided a detailed description of what it would take to run the world on renewable energy. By “run” they mean providing all energy for all purposes (electric power, transportation, and heating/cooling), everywhere in the world, from wind, water, and the Sun (WWS)—no fossil fuels, no nuclear power, just renewable power. Their report, “Providing all global energy with wind, water, and solar power,” is divided into two parts: “Part I: Technologies, energy resources, quantities and areas of infrastructure, and materials” and “Part II: Reliability, system and transmission costs, and policies.” In the first, they arrive at the estimated amounts and types of equipment needed to build what the authors refer to as a WWS energy infrastructure.

One scenario for powering the world with a WWS system includes 3.8 million 5 MW wind turbines (supplying 50% of projected total global power demand in 2030), 49,000 300 MW CSP power plants (supplying 20% of demand), 40,000 solar PV power plants (14%), 1.7 3 kW rooftop PV systems (6%), 5350 100 MW geothermal power plants (4%), 900 1300 MW hydroelectric power plants, of which 70% are already in place (4%),720,000 0.75 MW wave devices (1%), and 490,000 1 MW tidal turbines (1%).

In Part II, the authors discuss methods of accommodating the variability of WWS energy to ensure that power supply reliably matches demand. Palliative steps include interconnecting geographically dispersed resources, using hydroelectricity to fill in shortfalls, using demand-response management, storing electric power on site, over-sizing peak generation capacity and producing hydrogen with the excess, storing electric power in vehicle batteries, and forecasting weather to more accurately project energy supplies. In other words, they try to address all of the proposed ideas about making renewable energy work reliably in one paper.

Given such limited space, the authors are not able to go into great detail about smart grid technology, or the idea of using the batteries in electric cars to store power for the national grid (we'll all have electric automobiles by 2030, right?), or a dozen other green flights of fancy. Al and I found we could only present a much simplified description of the same technologies in The Energy Gap, and that was a 400 page book.

Delucchi and Jacobson then address the economics of WWS generation and transmission, the economics of WWS use in transportation, and policy measures needed to enhance the viability of a WWS system. Using many overly optimistic estimates regarding availability and the ability of wide spread power transmission grids to balance loads from fluctuating wind and solar power sources, they claim that WWS is a viable replacement for today's reliable baseload electrical system. They do recognize that the major problem with wind and solar are their capricious natures:

As a result, there will be times when a single installation cannot supply enough power to meet demand and when the installation can produce more power than is needed, which can be an economic waste of generating capacity (but see item E in the list below). However, there are at least seven ways to design and operate a WWS energy system so that it will reliably satisfy demand and not have a large amount of capacity that is rarely used: (A) interconnect geographically dispersed naturally variable energy sources (e.g., wind, solar, wave, and tidal), (B) use a nonvariable energy source, such as hydroelectric power, to fill temporary gaps between demand and wind or solar generation, (C) use ''smart’’ demand-response management to shift flexible loads to better match the availability of WWS power, (D) store electric power, at the site of generation, for later use, (E) over-size WWS peak generation capacity to minimize the times when available WWS power is less than demand and to provide spare power to produce hydrogen for flexible transportation and heat uses, (F) store electric power in electric-vehicle batteries, and (G) forecast the weather to plan for energy supply needs better.

If there is one thing an electric power grid cannot abide, it is unpredictable power delivery. For a grid to work efficiently and reliably, demand and supply must be kept in balance. Dismissing such considerations, the authors breezily come to the conclusion that the only things standing in the way of a bold new green world are people's attitudes about renewable energy and a lack of political will. As they concluded: “We find that the cost of energy in a 100% WWS will be similar to the cost today. We conclude that barriers to a 100% conversion to WWS power worldwide are primarily social and political, not technological or even economic.”

I have discussed the unreliability and variable nature of wind power and questions regarding its true ecological impact before on this blog. At least a dozen articles have addressed the problems with renewable energy—including land use, ecological impact and engineering requirements—so I will not dive into the mind-numbing detail required to refute the authors' assumptions item by item here. What I will do is address the economics of the situation.

Most of the financial figures given for renewable energy are carefully chosen to show green energy in a positive light. The facts are renewable energy is still much more expensive than conventional electrical generation. And to be accurate, government subsidies and grants cannot be used to discount the cost because, in the end, it is the total cost to society that counts. Whether a power company, the government or consumers pay it all costs the economy. Looking at generation costs without considering initial purchase, installation and integration costs are also misleading. Quite frankly, there is no way the cost of WWS would be similar to energy cost today based on initial purchase cost alone.

As can be seen from the description above, there is enough detail in these papers to put even the most enthusiastic energy policy wonk to sleep. That is why here, as in other articles referring to these findings, the requirements have been simplified to 4 million 5 MW wind turbines and 90,000 300 MW solar power plants, ignoring the remaining 16% coming from rooftop solar panels, geothermal, tidal, wave and other sources. Using just the two primary sources, wind and solar, some quick calculations put the enormity of any replacement program into perspective:

1. The cost of large commercial wind turbines varies from $1 to $2 million per MW of nameplate capacity. Turbines 2 MW in size cost roughly $2.8 million installed. Ballpark figures for a 5 MW wind farm would expect to cost in the region of $9.7-14 (€7-10) million, whether from a signal large turbine or a constellation of smaller units. The figure represents the total project cost and includes the feasibility studies, EIS and planning application, civil and electrical engineering works, grid connection costs. Let's call it $10 million per 5 MW installed. Calculating the total cost for world wind power:
   
   4,000,000 * $10,000,000 = $40 Trillion


2. The solar component calls for the use of industrial scale concentrating solar plants, the most cost efficient form of solar power. Abengoa Solar, a company currently constructing solar thermal plants, put the cost of a 300 MW plant at 1.2 billion euros in 2007. In 2009, the Arizona state government announced a 200 MW plant for 1 billion US dollars so let's split the difference and estimate $1.56 billion per plant. Calculating the total cost for world solar power:
   
   90,000 * $1,560,000,000 = $140 Trillion

So, for 84% of the needed capacity the initial purchase and installation cost is 12 times the total yearly GDP of the USA. For comparison, using a power factor of 35%, the wind plants could be replaced by 1,000 nuclear plants at $12 billion apiece (2 GW each), totaling $12 trillion. The solar plants would require 4500 nuclear plants to replace them. This would cost $54 trillion. Renewable is 270% more than nuclear. And this is using a nuclear critic approved cost of $12 million per plant.

Current manufacturers of nuclear power plants out side the US are quoting much lower cost per unit (and much faster construction times, but that's another matter). Westinghouse claims that its AP1000 will cost $1billion for a 1.15 GW plant. That would reduce the above estimates by a factor of 6. In other words, total nuclear cost would be $9 trillion, making the renewable “solution” 2000% (20x) more expensive.

It should also be noted that the above costs are without the necessary, continent spanning power grids needed to match spotty wind and solar power with demand. It has been estimated that to upgrade the US power grid to accommodate renewable energy sources will cost $2 trillion over the next 20 years. While a system using nuclear power will undoubtedly need to be expanded in the future, because nuclear is baseload power (i.e. steady), it would not require the extra expense of intermittent sources such as wind, solar or wave. If we use total population as an indication of demand, and hence grid infrastructure need, this adds another $45 trillion to the WWS requirements.

The total bill for WWS comes to around $225 trillion over the next 20 years. That is nearly the entire output of the world's largest economy every year for two decades. Greens will say that once the system has been converted the energy costs drop, after all wind and sunshine are free. True, but fuel costs for nuclear power are also very low, and $150 Trillion will by a lot of uranium and thorium. And we know nuclear power works safely and reliably, the same cannot be said of renewable power generation on the scale being proposed.

Aside from the mind-boggling cost, WWS, as proposed by Jacobson and Delucchi, requires new, unproven technologies, rapidly falling manufacturing prices, and international cooperation unheard of today. Given the havoc caused by natural gas supply interruptions caused by Russia, would any sovereign nation trust a power grid that spans three continents and thousands of miles? A power grid that could be disrupted by terrorists or maniacal despots anywhere along its major arteries? Any way you look at renewable energy, it makes little sense.

Perhaps the best way to look at running the world exclusively on renewable power is that it would cost $33,500 for every man, woman and child on Earth. People in developed nations might be willing to invest this much, but what of those living in under developed economies, where per capita yearly income can be less than $300? Nobody but deep green zealots would call this a reasonable deal. If you are interested in a workable plan using currently available technology, pick up a copy of The Energy Gap. The world's future energy needs can be met while reducing pollution and without bankrupting everyone on the planet—it just cannot be done using wind and solar energy.

Be safe, enjoy the interglacial and stay skeptical.

PS: Here is a little widget to help put energy production into perspective from OilPrice.com. Note that for geothermal, wind and solar the numbers are in millions of BTUs, not billions.