No doubt you’ve heard people speak of an energy transition from a fossil fuel-based society to one based on renewable energy–energy which by its very nature cannot run out. Here’s the short answer to why we need do it fast: climate change and fossil fuel depletion. And, here’s the short answer to why we’re way behind: History suggests that it can take up to 50 years to replace an existing energy infrastructure, and we don’t have that long.

Perhaps the most important thing that people don’t realize about building a renewable energy infrastructure is that most of the energy for building it will have to come from fossil fuels.

Currently, 84 percent of all the energy consumed worldwide is produced using fossil fuels–oil, natural gas and coal. Fossil fuels are therefore providing the lion’s share of power to the factories that make solar cells, wind turbines, geothermal equipment, hydroelectric generators, wave energy converters, and underwater tidal energy turbines.

Right now we are producing at or close to the maximum amount of energy we’ve ever produced from fossil fuels. But the emerging plateau in world oil production, concerns about the sustainability of coal production, and questionable claims about natural gas supplies are warnings that fossil fuels may not remain plentiful long enough to underwrite an uneven and loitering transition to a renewable energy society.

This is what’s been dubbed the rate-of-conversion problem. In a nutshell, is our rate of conversion away from fossil fuels fast enough so as to avoid an unexpected drop in total energy available to society? Will we be far enough along in that conversion when fossil fuel supplies begin to decline so that we won’t be forced into an energy austerity that could undermine the stability of our society?

The answer can’t be known. But the numbers are not reassuring. Based on data from the U.S. Energy Information Administration, it would take more than 70 years to replace the world’s current electrical generating capacity with renewables including hydroelectric, wind, solar, tidal, wave, geothermal, biomass and waste at the rate of installation seen from 2005 through 2009, the last years for which such data is available. And, that’s if worldwide generating capacity–which has been expanding at a 4 percent clip per year–is instead held steady.

This also doesn’t take into account the amount of energy actually produced versus what is called nameplate capacity. Nameplate capacity is what a wind generator could generate if it operated at maximum capacity 100 percent of the time. But in practice, the turbines are only spinning when the wind blows and then not always at the maximum speed. This so-called capacity factor was just 27 percent for wind farms in the United Kingdom from 2007 to 2011. For solar photovoltaic the number was 8.3 percent. Even hydroelectric stations ran at only about 35 percent of capacity. This compares to about 42 percent for conventional coal, 61 percent for natural gas, and 60 percent for nuclear power stations (PDF). The contrast is starker using U.S. numbers: 72 percent for coal and 91 percent for nuclear using 2008 figures, though natural gas was only 11 percent, probably because these were primarily plants that only come on to meet peak demand and so don’t run very often.

What this means is that installing two to three times our current nameplate capacity in the form of renewables may be required to replace existing fossil-fueled plants. So, the transition period would actually turn out to be longer than what I’ve calculated, perhaps 140 to 210 years using 2005 to 2009 installation figures.

Of course, installations of such renewables as wind and solar are accelerating. So, that would tend to shorten this longer transition period–as would leaving existing nuclear power capacity intact. But would we be able to shorten the transition period enough to head off declines in total energy production and prevent additional serious damage to the climate?

Of course, some would say that we need to expand nuclear power generation rapidly to meet these challenges. Whether you support such an expansion or not, there are three key problems. First, building enough nuclear power stations to replace fossil fuel-fired plants would be the largest construction project ever undertaken and require the use of enormous amounts of fossil fuels. Making the necessary concrete alone would be a large new contributor to greenhouse gas emissions. That means that the initial phase of a nuclear transition would actually increase the rate of fossil fuel emissions. The savings on fuel and emissions wouldn’t come until much later.

Second, after the Fukushima disaster, there doesn’t seem to be much appetite for such a buildout. I’ll be very surprised if nuclear power generation even maintains its current level in the next 20 years as Japan and Germany abandon nuclear power. Third, the timeline for such a buildout would be measured in decades, partly because of the sheer logistics involved and partly because of the brake that regulatory approvals put on such projects. Even new, cheaper and easier-to-build designs may not help if they cannot achieve the necessary regulatory approvals promptly. The history of such approvals is not encouraging. The safest thing a nuclear regulatory agency can do is say no.

I haven’t even touched on replacing the fuels which power our transportation system and provide heat for our buildings and industrial processes. Transportation offers an extraordinary challenge since 80 percent of all transportation fuel worldwide is still derived from petroleum. In the United States the number is 93 percent. Despite billions of dollars spent and decades of research, we still have no good substitutes that scale to the size necessary to replace petroleum for transportation fuel.

Biofuels offer little hope. Already the ethanol bubble has burst. Biofuels–today mainly ethanol and biodiesel–compete with food. There is simply not a limitless supply of suitable farmland, and so there will be competition with the demand for food until we find substitutes for the industry’s main feedstocks, namely corn, sugar and soybeans.

Beyond this the problem of scale is simply unsolvable. To supply the entire U.S. car fleet–assuming it could run on ethanol–we’d have to plant 1.8 billion acres in corn for ethanol continuously. There are only about 440 million acres in the United States in cultivation now. And, it’s worth noting that current methods of corn cultivation require the copious use of herbicides and pesticides made from oil; tractors and other vehicles that run on oil to plow, harvest and spray the fields as well as transport the crop; and natural gas-derived nitrogen fertilizers to boost growth and replenish depleted soil. Fossil fuels are currently integral to growing corn, and I cannot see the wisdom of growing organic corn for anything but food.

As for heat for buildings, certainly we could insulate and seal our existing buildings better. And, this points the way to achieving an energy transition within the time we need to achieve it. Since it will probably be impossible to scale renewable energy fast enough to a level sufficient to produce the amount of energy we use today, the one absolute necessity to a successful energy transition is reducing consumption drastically. No politician dares to say anything remotely approaching this. And yet, it would be the cheapest, fastest way to address the twin crises of fossil fuel depletion and climate change.

Now, when I say reduce, I mean on the order of 80 percent over the next 20 to 30 years. For Americans this may seem impossible until they contemplate that the average European lives on half the energy of the average American. So often we hope for technological breakthroughs that will give us all the clean energy we desire. But we ought to focus equally, if not more, on using our prowess to find ways to reduce our energy consumption drastically. This is actually the much easier road. When we are made conscious of our energy use, we can change our behavior quickly to modify it without compromising the quality of our lives. As more homes and businesses are given the means to monitor their energy use, the people in them will change to lower their consumption and costs.

Already we know how to build so-called passive design structures which can lower energy use by 80%. And, we desperately need to figure out how to apply these techniques cheaply and economically to existing homes and businesses. In transportation we need to stop thinking that cars equal transportation and instead realize that cars provide the service of transportation which can be obtained in a number of ways, many of which use much less energy.
We may also need to speed the energy transition in electric power generation using so-called feed-in tariffs. These tariffs–which harness the ingenuity of countless small producers–have enabled Germany to expand solar, wind and other alternatives so that they generate 25 percent of its electricity today. Germany, not a particularly sunny place, is currently the world’s top generator of solar electricity.

Of course, per person energy consumption in poor countries is only a small fraction of that in rich countries. We cannot expect the world’s poor to reduce their energy use by 80 percent. Instead, we must help them to move quickly beyond fossil fuels to renewable energy.

By simultaneously reducing consumption and encouraging a rapid buildout of renewable energy, it is possible that we could mitigate the problem of declining fossil fuel supplies before it becomes so acute that it would cripple that very build out. And, we could address climate change at the same time. Certainly, there are difficult problems to be solved with renewable energy, storage being the key one. Most renewable energy comes in the form of electricity, and since there is often a mismatch between the time we produce that electricity and the time we need it, we will have to master storage.

But we will need a lot less storage if we focus on reducing consumption. This is the one strategy which will allow us to overcome the rate-of-conversion problem and achieve an energy transition in far less time than we have in the past.