NASA's Mod-1 wind turbine standing on a hilltop at Boone, North Carolina, in 1979.
Green Initiative

Energy sources compared

Every kilowatt-hour has a hidden history: a mine, a factory, a chimney or a blade. Compare them honestly and the picture is clearer than the debate.

NASA Glenn Research Center

In 1979 NASA switched on the Mod-1, a two-bladed wind turbine with a rotor 61 metres across, on a mountain in North Carolina. It was one of the largest turbines of its day and one of the first attempts to make wind power a serious part of the grid. Four decades later wind and solar are the fastest-growing sources of electricity on the planet. To understand why, and what they still cannot do, it helps to compare every source on the same terms.

The whole life, not just the chimney

A power station’s emissions do not stop at its chimney. Steel and concrete must be made, fuel mined and transported, and everything eventually dismantled. Lifecycle assessments add all of this up per kilowatt-hour. The IPCC’s review of hundreds of such studies gives a clear ranking: coal emits around 820 grams of CO2-equivalent per kilowatt-hour, gas about 490, while wind, nuclear and hydropower are around 10 to 25, and solar around 40 to 50.

Safety tells a similar story. Counting deaths from accidents and from air pollution per unit of electricity, coal and oil are the most dangerous sources by far, because burning them fills the air with particles. Wind, nuclear and solar are hundreds of times safer.

820 gCO2-equivalent per kWh from coal, lifecycle median (IPCC AR5)
11 gCO2-equivalent per kWh from onshore wind (IPCC AR5)
24.6deaths per TWh from coal, against 0.04 for wind (Our World in Data)

What each source is good at

  • Coal and gas can be switched on when needed, but carry the highest emissions and pollution. Gas emits about half as much CO2 as coal when burned, though methane leaks narrow the gap.
  • Nuclear runs steadily with very low emissions and a tiny land footprint; it is expensive and slow to build, and waste must be managed for a very long time.
  • Hydropower is flexible and low-carbon, but dams flood valleys and change rivers.
  • Wind is among the cheapest new electricity; it varies with the weather and needs space, grids and, increasingly, storage.
  • Solar is cheap, modular and fits on rooftops from a village to a skyscraper; it produces nothing at night and less in winter.
Two giant bucket-wheel excavators in a vast sandy open-cast mine, with green fields and a row of wind turbines on the horizon.
Bucket-wheel excavators in the Garzweiler open-cast lignite mine in western Germany, 2019, with wind turbines on the horizon. For coal, the hidden history of a kilowatt-hour starts in a mine like this. Photo: Arne Müseler, CC BY-SA 3.0 DE

There is no perfect source. There are only better and worse mixes, and the mix is ours to choose.

The missing piece: flexibility

The challenge of a clean power system is not generating enough energy on average, but matching supply and demand every second. Batteries, pumped hydro, stronger grids that share power across regions, demand that follows the sun and wind, and firm low-carbon sources all fill that gap. This is also why we research modular generation that can sit where power is used, such as the Fan Wall.

Proven solutions: grids that already changed

Clean electricity is not only a forecast. Several countries have rebuilt most of their power supply within a generation, each with the resources it had: wind, water, the heat of the Earth or the sun. All figures below are shares of each country’s electricity generation, from Ember and the Energy Institute as compiled by Our World in Data, unless another source is given.

  • Denmark: wind as the backbone. Fossil fuels made 84 percent of Denmark’s electricity in 2000 and 11 percent in 2024. Wind turbines alone made 58 percent, bioenergy about a fifth, and solar has grown from almost nothing to 11 percent since 2012.
  • Uruguay: a transition in less than a decade. Uruguay used to rely on its dams and burn oil and gas in dry years; in 2008, fossil fuels made 39 percent of its electricity. Wind grew from about 1 percent in 2013 to 32 percent in 2018. Even in 2023, a year of very low hydro output, fossil fuels made under 8 percent, and in 2024 about 1 percent.
  • Kenya: heat from the Rift Valley. Geothermal power, from steam wells in the Rift Valley, made 10 percent of Kenya’s electricity in 2000 and 45 percent in 2024, more than any other source. With hydro, wind and solar, more than nine tenths of Kenya’s electricity is now renewable, up from 40 percent in 2000.
  • Scotland: more than it needs. Scotland generated a record 38.4 TWh of renewable electricity in 2024, 30.1 TWh of it from wind, and exported 19.7 TWh more electricity to the rest of the United Kingdom than it imported (Scottish Government, 2025).
  • The United Kingdom: the end of coal. In 2000 three quarters of British electricity came from fossil fuels; in 2024 about a third did, and wind made 29 percent. In autumn 2024 Ratcliffe-on-Soar, the last coal power station, closed after 50 years, making Britain the first major economy to stop burning coal for power (UK Government, 2024).
  • Brazil: water, then wind and sun. Hydropower has supplied most of Brazil’s electricity for decades. Since 2010 the country has added wind and solar quickly: together they made 24 percent of its electricity in 2024, up from less than half a percent in 2010.
  • Costa Rica: almost all renewable. Hydro, geothermal and wind have supplied more than 98 percent of Costa Rica’s electricity in every year since 2015.
  • Bangladesh: light without a grid. Many villages in Bangladesh were far from any power line. From January 2003 the Bangladeshi finance company IDCOL, first backed by the World Bank, financed small solar home systems for off-grid rural households. By January 2019 about 4.1 million had been installed, bringing solar electricity to about 18 million people, 12 percent of the population, who had lit their homes with kerosene lamps (IDCOL).
White steam rising from a long low geothermal power plant on a dry hillside under a cloudy sky.
The Olkaria III geothermal power plant near Naivasha, in Kenya's Rift Valley. Steam from wells like these now makes more of Kenya's electricity than any other source. Photo: Bbossoxx, CC BY-SA 4.0

Why it got cheap: the learning curve

The reason these transitions sped up is price. A solar module cost about 132 dollars per watt in 1975 and 27 cents in 2024, in today’s money: 99.8 percent cheaper (Our World in Data). Economists call the pattern a learning curve: every time the world’s total installed capacity has doubled, the price of a module has fallen by roughly a fifth to a quarter, as factories grew, designs improved and supply chains matured. Batteries followed the same path. Lithium-ion cells cost around 9,200 dollars per kilowatt-hour in 1991 and 78 dollars in 2024, more than 99 percent less (Ritchie, Our World in Data), and the average price of a battery pack fell by 20 percent in 2024 alone, to 115 dollars per kilowatt-hour (BloombergNEF, 2024).

The cheapest energy of the future is the energy we practise building. Every panel installed makes the next one cheaper.

Beat the benchmark: a call for ideas

The comparison above has a clear lesson: any new way of making or saving energy has to compete with sources that are already cheap and clean. On a building, the benchmark is a rooftop solar panel. That is a high bar, and it is exactly why we want more people thinking about it. Our own Fan Wall is one idea we are testing openly, including where it fails. It is only one idea. The real task is to find solutions to the problems that remain: power at night and in winter, heat for homes and industry, crowded roofs, and grids that need flexibility.

An aerial view of a suburb of red and grey roofs among gardens and trees, many of the roofs covered with dark solar panels.
Suburban roofs with solar panels in Canberra, Australia. On a building, this is the benchmark every new idea has to beat. Photo: Benlisquare, CC BY-SA 4.0

We would like to hear your ideas, and there are two ways an idea can go from here.

  • A pitch needs a working model. Some members of the Foundation may have contacts who could hear a pitch for a strong idea. To be pitched, an idea needs a working, real-life model that actually produces power, with its output measured on an instrument, not claimed. A kettle-sized prototype with a logged meter says more than any drawing.
  • Paper ideas are welcome too. A well-argued idea without hardware can still be published here on the website if it is solid: the physics checks out, the numbers are honest and the cheapest test is clear. That is how good ideas find the people who can build them.

Be realistic about the difference: raising capital without a working model is very difficult, and nothing here is a promise of funding, returns or introductions. What we can promise is a serious, public reading of the numbers.

Every alternative, explained

Each article covers the physics, the hard limits, real projects with measured results, the cost against rooftop solar, and what a working prototype would need.

Directions worth exploring

These are starting points with real physics behind them, each with the limit an idea has to respect. Every idea on the radar has its own in-depth article, linked below, written with the same honesty as our Fan Wall research.

  • Solar surfaces that do two jobs. Facades, noise barriers, car-park canopies and canal covers that make electricity while shading or protecting something else. Hybrid PV-thermal panels add heat: roughly 100 to 700 kWh of heat per square metre a year on top of the electricity (IEA SHC Task 60, 2020). The limit is sunlight itself: about 1 kW per square metre at best. Read: solar roofs that also make heat.
  • Farms and reservoirs that also harvest sunlight. Agrivoltaic systems have been modelled to get 35 to 73 percent more from the same land than separate crops and panels (Dupraz et al., 2011). Floating solar on 30 percent of suitable reservoirs could produce about 9,400 TWh a year, roughly a third of today’s world electricity, and cut evaporation (Jin et al., 2023). Read: agrivoltaics and floating solar.
  • Heat, not just electricity. Heat makes up about half of the world’s final energy use (IEA, Renewables 2019). Sand and stone heat batteries store cheap wind and solar power for weeks; a 100 MWh unit began heating the Finnish town of Pornainen in 2025 (Polar Night Energy). Ground-source heat pumps deliver about four units of heat for each unit of electricity (IEA, 2022). Waste heat from factories, sewers and data centres is still mostly thrown away (Forman et al., 2016). Read: sand and heat batteries, ground-source heat pumps and waste heat.
  • Water that is already moving. Tidal currents are predictable decades ahead; the MeyGen array in Scotland passed 50 GWh in 2023. Waves, rivers, irrigation canals and pressure-reducing valves in water mains all carry energy, usually only kilowatts per site, which is why cheap, standard designs matter more than record-breaking ones. Read: tidal stream, wave energy and small and in-pipe hydro.
  • Higher winds. Wind power grows with the cube of wind speed, and winds a few hundred metres up are stronger and steadier. Energy kites are the attempt to reach them; Google’s Makani closed in 2020 after thirteen years, while smaller companies keep testing. A good idea here must explain reliability, not just power. Read: energy kites.
  • Using energy at the right time. Shifting when we heat water, charge cars and cool buildings is the cheapest “power station” of all: the United States alone has about 200 GW of cost-effective flexible demand by 2030 (Brattle, 2019). Ideas in software, tariffs and simple home devices count as energy ideas too. Read: demand flexibility.
An aerial view of a multi-storey car park whose top floor is covered by long rows of solar panels on steel canopies, among trees.
Solar canopies over the top floor of a car park in Natick, Massachusetts: the same surface shades the cars and makes electricity. Photo: Stephen Coffrin, US Department of Energy (public domain)

What a strong idea brief contains

  1. The problem it solves, and for whom: a city roof, a farm, a factory, a village without a grid.
  2. The physics check. How many watts per square metre, or per tonne, or per litre of water? How many kWh a year? A quick comparison with the benchmark above saves everyone time; David MacKay’s Sustainable Energy Without the Hot Air is a free guide to doing this honestly.
  3. The cost per kWh, even as a rough range, and what would have to become cheaper for it to win.
  4. What would have to be true, and the cheapest experiment that could show it is not.
  5. Who it could harm, from wildlife to neighbours, and how you would know.

A good idea does not need to be right. It needs to be testable, and honest about what would prove it wrong.

Organisations doing this work

These non-profit organisations help bring clean, affordable energy to people and publish what they achieve. Have a look at their work and, if it convinces you, consider giving to them directly. The Local Solar System Foundation has no affiliation with any of them and receives nothing from them. Before you give, read each organisation’s own reports and accounts and check how it measures its results. The figures below are the organisations’ own.

  • SolarAid (UK registered charity 1115960) brings solar lights to rural families in sub-Saharan Africa who would otherwise use kerosene, candles or torches, through its own social enterprise. It reports 2.5 million lights distributed and 13.3 million people reached.
  • Solar Sister (US non-profit) recruits, trains and supports women entrepreneurs in remote parts of sub-Saharan Africa to build businesses selling solar lighting and clean cookstoves to households off the grid.
  • Practical Action (UK registered charity 247257) works with communities in Africa, Asia and Latin America on practical technology against poverty, including access to clean energy, for example renewable energy for refugee camps.
  • RMI (founded in 1982 as Rocky Mountain Institute, US non-profit) is an independent research organisation that works with governments, power companies and businesses in more than 50 countries to move energy systems to clean, low-cost power.
  • Ember (independent not-for-profit think tank, founded 2020) publishes free, open data on the electricity of almost every country, including much of the data in the chart above, so that anyone can check how fast grids are changing.
A young girl in a school pinafore looks down at a small yellow sunflower-shaped solar lamp glowing in her hands.
In Addis Ababa, Ethiopia, a schoolgirl holds her first solar-powered lamp. For families without a grid connection, a small solar light can replace kerosene and candles. Photo: USAID in Africa (public domain)

How you can help

  • Switch your supply to a renewable tariff or community energy scheme where one is available.
  • Use power when it is clean. Run washing machines and charge vehicles when wind and solar are plentiful; many grids publish real-time carbon intensity.
  • Put panels on roofs. Homes, schools, supermarkets and car parks are ready-made solar sites; ask your council and your employer.
  • Save before you generate. Insulation and efficient appliances cut bills and emissions at the same time.
  • Ask for the numbers. Ask your electricity supplier and your government for the lifecycle emissions of the power you buy.
  • Support the people doing the work. The organisations above bring solar light and clean power to families who have none, and publish open data and research; check their reports and give directly if their work convinces you.
  • Send us an energy idea. Use the contribution form, choose the Green Initiative, and describe your idea with the five points above. If you have a working model, include its measured output and how you measured it. Engineers, students and makers are also welcome to review other people’s ideas on the contribution board.

Sources and further reading