A forecast this month showed the world might need 124 percent more energy in 2100, raising questions over where the capacity will come from.

Dr. Euan Mearns, writing on the Energy Matters blog, looked at the expected increase in population and per capita energy consumption between 2015 and the end of this century, and concluded annual demand could top 29.5 billion tonnes of oil equivalent (TOE).

This is roughly equal to 343 petawatt-hours, or around 54 times the amount of all the renewable energy produced in the world in 2017, based on International Energy Agency data.

The forecast figure reflects a United Nations medium population growth prediction which would see 11.2 billion people inhabiting the planet by 2100.

Per capita energy consumption, meanwhile, is expected to grow in a linear fashion to 2.6 TOE per person a year in 2100, a 47 percent increase on 2015 levels, following the trend seen since 1965.

The trend, based on data from the 2016 BP Statistical Review of World Energy, reflects the fact that per capita energy consumption is rising in countries such as China, India and Brazil, even as it begins to fall in many OPEC markets.

Under a low United Nations population growth forecast, the world would still end up consuming 19.16 billion TOE, or about 223 petawatt-hours a year, up 46 percent from 13.15 TOE or around 153 petawatt-hours in 2015.

“I have not included the high forecast because it produces silly demands on Earth resources,” commented Mearns in his analysis.

It is unclear how society will cope with the energy demands of even a medium forecast in population. BP, for example, sees natural gas and crude oil running out in the 2060s

Even the longevity of coal, which is expected to last well into the next century, could be overstated, according to production curve studies by David Rutledge, Kiyo and Eiko Tomiyasu Professor Emeritus of Engineering at the California Institute of Technology.

Rutledge told GTM he expected coal to stop playing a major role in the global energy mix within 60 years. “Many of us have felt that coal production will peak when China peaks,” he said. “It may already have happened.”

Nuclear power could step in. China is set to lean heavily on the technology, with plans to triple its nuclear generating capacity in the next two decades. Elsewhere, though, nuclear has lost its shine.

In the U.S., for example, nuclear plants are struggling to remain commercially viable despite the administration’s haphazard attempts to support the industry. In Europe, even France is backtracking on nuclear. The outlook for new generation in the U.K. remains uncertain.

Nuclear proponents point to new technologies, from small modular reactors  to thorium plants, which could potentially overcome many of the disadvantages of today’s designs. But bringing these novel reactor ideas to market is not proving easy. That leaves renewable energy and efficiency.

In BP's 2018 Energy Outlook, the oil and gas giant predicted -- for the first time -- that liquid fuel demand (not supply) would peak in the late 2030s. "Growth of fuels used in transport slows as the impact of rising prosperity is offset by efficiency gains," the report states. Encouraging greater efficiency across the entire energy sector would help to mitigate unsustainable demand, but wouldn't entirely solve the problem.

Renewables are the other lever society can pull.

Stanford University’s Professor Mark Jacobson has famously claimed the world could be 100 percent powered by wind, water and sunlight by 2050, but it is not clear how his models might scale to account for population and per capita energy growth by the end of the century.

Furthermore, Jacobson’s work remains controversial. Last year, Jacobson filed a $10 million lawsuit (recently dropped) against the Proceedings of the National Academy of Sciences and grid researcher Chris Clack over a paper that called Jacobson’s work “a poorly executed exploration of an interesting hypothesis."

Concerns with 100 percent renewable models tend to focus on monetary cost and how to deal with intermittency. But an equally big challenge might be renewable energy’s energy return on investment (EROI).

This is the amount of energy gained from an energy production technology relative to the energy that must be used to get it.

A very low EROI means it is only marginally worthwhile to develop a technology, and below a given level you may not get enough surplus back to sustain society. Hydroelectric power has the highest known EROI, at 84:1, albeit with large variations depending on the site.

Wind scores a respectable 18:1 and solar comes in at 10:1. But the EROIs of intermittent renewables suffer heavily if they have to be tied to battery storage or backup systems to provide a reliable supply.

All of this may mean society will have to go through some serious readjustments this century, said Charles Hall, author of Energy and the Wealth of Nations and the founding father of the EROI concept. “What’s the future? I don’t think it’s going to be like it is now,” Hall said.

“Can we build a good structure on renewable energy? I think perhaps. But my economics co-author, Kent Klitgaard, says, ‘You can’t have capitalism,’ because capitalism requires growth. As soon as you bring that up, you open a whole new can of worms.”

This story was updated to include findings from BP's 2018 Energy Outlook.