Power density: why land and space matter: What power density is
Understanding Energy / Module 8: Power density: why land and space matter Part 1 of 5What power density is
Energy debates usually focus on cost and capacity. Smil argues that a third measure, the rate of energy flow per unit of area, is just as basic and far more often ignored. It explains why a coal mine is small and a wind farm large, why cities depend on imports, and why switching energy sources changes landscapes.
1.1 Power density is watts per square metre, and the details matter
Power density is power divided by area. Its unit is the watt per square metre, W/m². It answers a simple question: for every square metre of land or surface involved, how fast does energy flow? The idea can describe a resource, such as sunlight falling on a field; a production process, such as a coal mine or a wind farm; or a use, such as a house, a steel mill or a whole city.
The definition is simple, but a figure means nothing until two choices are stated. The first is which power. Is it the peak rate, for example a solar panel at noon in July, or the average over a whole year including nights and winter? The second is which area. Is it only the area of the solar panels, or the whole fenced site including gaps between rows, roads and substations? For a wind farm, is it the concrete pads and access tracks, or the whole area across which turbines must be spaced so they do not steal each other's wind?
Smil stresses that these choices can change the answer by ten times or more. A solar module produces about 200 W per square metre of its own surface at noon. Averaged over a year, and spread over the whole area of a solar park, the same technology delivers only about 4 to 11 W/m². Neither number is wrong, but comparing one source's peak panel figure with another's annual site average is meaningless.
Unless stated otherwise, figures are annual averages over the whole area a facility claims. That is the most honest basis for asking how much land a society needs to supply its energy. A further subtlety is time. A mine disturbs land for decades and may later be restored; a solar park occupies land for 25 to 40 years but leaves it largely intact; a reservoir may flood land permanently. Power density captures how much land is tied up per unit of energy flow, not how badly or how permanently it is changed.
- Smil, Power Density: A Key to Understanding Energy Sources and Uses (2015) · via Smil Mastery Course, M8 'What power density is and why it matters'
- Smil, Energy in Nature and Society: General Energetics of Complex Systems (2008) · via Smil Mastery Course, M8 'What power density is and why it matters'
1.2 The area an energy system occupies shapes its whole structure
Economists judge energy sources by price, and engineers by capacity and efficiency. Smil's case is that land claims belong beside them, because the space a system needs predetermines much of its infrastructure, cost and environmental footprint. He leans on the energy analysts Wolf Häfele and Wolfgang Sassin, who ranked the spatial density of energy operations among the most important things to know about any energy system.
Consider what a high power density buys. A large coal mine, an oil field or a gas field delivers vast flows of energy from a small patch of ground. A small number of sites can then supply a whole country through pipelines, railways and power lines. The industry is concentrated, the land disturbed is modest, and the cities and factories it serves can be far larger than the area of production.
A low power density forces the opposite arrangement. If a source delivers only a few watts per square metre, gathering a large flow of energy means covering a large area with collectors, whether panels, turbines or crops, and then building a dense web of cables or roads to collect that energy and move it to where it is used. The collecting equipment itself becomes the main cost, the materials needed per unit of energy rise, and the energy system becomes visible across whole landscapes rather than at a few industrial sites.
Not everyone accepts that this matters. The efficiency advocate Amory Lovins once dismissed land footprint as an odd criterion, arguing that land under wind turbines can still be farmed and that rooftops cost no extra land. Smil's reply is that both points are partly true and still miss the main issue: even with shared use, the number of machines, the length of lines and the materials needed all scale with the area that must be harvested.
Power density does not decide whether a source is good or bad. It tells you what kind of system you will have to build to use it, and how much of the landscape that system will touch.
- Smil, Power Density: A Key to Understanding Energy Sources and Uses (2015) · via Smil Mastery Course, M8 'What power density is and why it matters'
- Smil, How the World Really Works: A Scientist's Guide to Our Past, Present and Future (2022) · via Smil Mastery Course, M8 'What power density is and why it matters'
1.3 Land-use intensity is power density turned upside down
Much recent research does not quote watts per square metre. Instead it reports land-use intensity: how many hectares are needed to produce one terawatt-hour of electricity each year, written ha/TWh/y. The two measures carry the same information, one being the inverse of the other, and it pays to be able to switch between them.
The conversion is a single number. One terawatt-hour a year is a steady average flow of about 114 MW, because a year has 8,760 hours. One hectare is 10,000 square metres. So power density in W/m² equals roughly 11,400 divided by the land-use intensity in ha/TWh/y. A source that needs 1,000 hectares for each TWh a year runs at about 11 W/m²; one that needs 10,000 hectares runs at about 1.1 W/m².
A second useful habit is to convert a power density into the area needed for a given demand. Divide the average power you need by the power density. Supplying an average of 1 GW, about 8.8 TWh a year, at 10 W/m² needs 100 million square metres, or 100 km². At 1 W/m² it needs 1,000 km², and at 0.2 W/m² it needs 5,000 km².
These conversions bring abstract figures down to earth. A city of 1,500 km², roughly the size of Delhi's national capital territory, would need its whole area covered by solar parks to supply about 15 GW on average at 10 W/m². India's whole electricity demand in 2025, around 2,100 TWh, corresponds to an average flow of about 240 GW.
One caution applies. Land-use studies usually express output as electricity, while Smil often quotes fossil fuel extraction in terms of the heat content of the fuel. A coal mine's thermal power density is about three times the electrical power density of the coal plants it feeds, because only about a third of the coal's heat becomes electricity. Always check which kind of energy a figure measures.
| Land-use intensity (ha per TWh a year) | Power density (W/m2) | Area for 1 GW average output (km2) |
|---|---|---|
| 10 | about 1,140 | about 0.9 |
| 100 | about 114 | about 9 |
| 1,000 | about 11 | about 88 |
| 10,000 | about 1.1 | about 880 |
| 50,000 | about 0.23 | about 4,400 |
Power density = 11,416 / land-use intensity. 1 GW average = 8.76 TWh a year.
Source: Lovering, Swain, Blomqvist and Hernandez, Land-use intensity of electricity production and tomorrow's energy landscape, PLOS ONE 17(7): e0270155 (2022), unit definitions, https://doi.org/10.1371/journal.pone.0270155; conversion is simple arithmetic- Lovering, Swain, Blomqvist and Hernandez, Land-use intensity of electricity production and tomorrow's energy landscape, PLOS ONE 17(7): e0270155 (2022) · web
- Ember, Global Electricity Review 2026 (April 2026) · India demand change 2025, https://ember-energy.org/latest-insights/global-electricity-review-2026/2025-in-review/
See it in the table
St SteelCheck your understanding
A study reports that a technology needs 2,000 hectares per TWh of electricity a year. What is its power density?
Divide 11,400 by 2,000 to get about 5.7 W/m².Which comparison is fair?
Only like-for-like comparisons make sense: same type of power (annual average) over the same kind of area (whole site).How much land does a source at 1 W/m² need to supply 1 GW on average?
1 GW is a billion watts; at 1 W per square metre that needs a billion square metres, which is 1,000 km².