NREL map

It Takes 2.8 Acres of Land to Generate 1GWh of Solar Energy Per Year, Says NREL

NREL mapThe Energy Department’s National Renewable Energy Laboratory (NREL) has published a report on the land use requirements of solar power plants based on actual land-use practices from existing solar facilities.

The report, Land-use Requirements for Solar Power Plants in the United States, shows results from data gathered from 72 percent of the solar power plants installed or under construction in the United States. Among the findings:

  • A large fixed tilt photovoltaic plant that generates 1 GWh per year requires, on average, 2.8 acres for the solar panels. This means that a solar power plant that provides all of the electricity for 1,000 homes would require 32 acres of land.
  • Small single-axis PV systems require on average 2.9 acres per annual GWh – or 3.8 acres when considering all unused area that falls inside the project boundary.
  • Concentrating solar power plants require on average 2.7 acres for solar collectors and other equipment per annual GWh; 3.5 acres for all land enclosed within the project boundary.

By the third quarter of 2012, the United States had deployed more than 2.1 GW of utility-scale solar generation capacity. Another 4.6 GW was under construction.

The report authors found that many of the solar land-use ranges and estimates used in anecdotal literature are very close to actual solar land use requirements. These land-use estimates can also be compared with other energy-production land uses. For example, a study by Vasilis Fthenakis and Hung Chul Kim of Columbia University (2009) found that, on a life-cycle electricity-output basis – including direct and indirect land transformation – utility-scale PV in the Southwest requires less land than the average US power plant using surface-mined coal.

A previous NREL report, “Land-use Requirements and the Per-capita Solar Footprint for Photovoltaic Generation in the United States,” had estimated that if solar energy was to meet 100 percent of all electricity demand in the US, it would take up 0.6 percent of the total area in the country.

GHS Label Guide
Sponsored By: VelocityEHS

Leveraging EHS Software in Support of Culture Changes
Sponsored By: VelocityEHS

Four Key Questions to Ask Before Your Next Energy Purchase
Sponsored By: EnerNOC, Inc.

Practical Guide to Transforming Energy Data into Better Buildings
Sponsored By: Lucid


15 thoughts on “It Takes 2.8 Acres of Land to Generate 1GWh of Solar Energy Per Year, Says NREL

  1. We shouldn’t compare this to coal or gas plants. We should compare this to the amount of acreage permanently lost under hydro. This beats the everloving pants off of typical hydro.

    In Alabama, the average acres of full reservoir per MW of hydro nampelate is 112 acres/MW.

    NREL says 2.8 acres can make 1GWh/year, which is 114kW. That’s derated power, not nameplate. That’s 24 acres/MW. Nameplate capacity would be 4 acres/MW!

    Compare 4 acres/MW for solar to 112 acres/MW for hydroelectric. If new hydro can be built where possible because it doesn’t pollute like coal and gas, then it looks like we should be rolling out solar with wild abandon.

  2. Also we should be looking at building more solar on the roofs of commercial buildings, so much space is already available. The solar panels will help keep rain and the sun off the normal building roof which will increase the building lifespan as well.

  3. There are lots of benefits that come from hydro that you don’t get with solar or wind. The watersheds and lakes used for recreation have a tremendous value to society.

    • Agreed, David. There are several reasons why we need reservoirs.

      1) We need the flood control.

      2) We need the water supply for useage.

      3) We need the lakes for recreation.

      4) We need the power generation.

      Dams and reservoirs are their own justification. The power is a bonus. They don’t need to be compared to anything else to prove their value and utility to society.

  4. Solar power plants remain the most versatile, easily-deployed and environmentally friendly of all energy sources so far.

  5. Try this…a footprint of 200 metres by 150 metres, 60,000 m, 335 megawatts 100 year timeframe, zero increase in reservoir size.

    To drag the size of the reservoir into the equation, you must also include the impact of mining the metals and rare earths elements required for those solar panels.

    It all depends on geography. Solar here in most parts of Canada makes little sense for REAL power for an energy grubbing population.

  6. We have acreage smack in between a large solar farm and wind energy farms in Southern California. There is legal access to the land. We’re now looking for an energy consultant and a feasibility study.

  7. In Sarnia Ont. Canada where I live 500 acres of the best farmland in Canada has been converted to solar. The loss of green space which actually removes CO2 combined with the loss of food production, at a subsidized price of 82 cents/kwh (6 cents for nuclear in Ont.) causes more harm to the low income families and actually worsens green house gases as compared to nuclear.

    • No question about it, Peter.

      Anyone who wants to deploy solar power in Canada is either a complete blithering idiot or an extremely slow learner who needs to spend about a year staring at a solar resource map.

      The same goes for the UN-analytical MORONS in Germany who who’ve installed enough nameplate solar capacity to supply 60% of Germany’s electricity needs, but which actually supplies only 5% of their needs BECAUSE THEY LIVE SO DAMN FAR NORTH THAT THEIR PANELS ARE RARELY ILLUMINATED.

      The pseudo-environmental religious Left needs to figure out what the words “mathematical analysis” and “return on investment” and “economizing” and “efficient use of resources” all mean. They also need to figure out that nuclear power is a critical and essential part of our future energy mix . . . especially if and when we start charging 100 to 300 million electric vehicles. That is going to double our demand for electrical generation.

  8. So according to the US uses max 966 GW total highest peak over the year. So if we were able to store energy in battery facilities- large batteries. Then I am supposing that we would need twice the energy to store from the day for the night and for the day. That means according to this we would need 76800 acres of land for solar or .003% of the land in the US to power the US for peak. The rest of the extra energy can be stored. With increasing inefficiencies of home and appliance design we may absorb any population increases.

  9. I think its interesting to notice that based on the numbers they give on average 0.032 acres (1394 ft^2) of solar panels provide enough energy for a home. That means you could just build the panels on top of already existing buildings, both single residence homes, apartment buildings, and other buildings to make up for the high density housing, and have enough energy to power all the residences in the country, and most likely enough left over to power much of the industry without any solar or wind farms.

    • Mr. Bean — wrong, wrong and wrong.

      The calculation may be that it takes 1400 sqft of panels to power one average home, but that’s under IDEAL solar farm conditions, such as . . .

      #1) Southward facing. The average home does not have a roof comprised of one single surface that’s conveniently facing due south. The average home (or apartment building) has roof surfaces that face several different directions — and only the ONE surface facing close to due south is usable. The others are useless.

      #2) That one surface represents only a portion of the total roof area. Also, many homes have exactly ZERO roof surfaces that are usable for solar. So only about 30% of the total roof area of American homes is useful for a solar installation.

      #3) Of that 30%, only about 70% can be solar panel area because there has to be enough clear roof around the panels to stand and work. 70% of 30% is 21%. So about 21% of the total roof area of American homes can be solar panels.

      #4) Rooftop solar installations are almost invariably fixed, with no tilting, so they don’t produce as much power as the ideal, ground-based, two-axis solar plants.

      #6) The figure of 32 acres to power 1000 homes — and your derived figure of 1400 sqft to power one home — are based on ideal irradiation conditions of the southwest American lattitude and extremely rare cloud cover. Conditions which are found only from Texas to southern California. About 80% of American homes lie outside of these conditions. You can install solar panels on the appropriate 21% of the roofs of that 80% of homes, but the power generation will be far less than is accounted for in the article’s figures.

      What’s more is that commercial and industrial electricity use is far greater than home use. BUT — if we install solar on all reasonable home and commercial and industrial roofs, that could probably produce 5% of our electrical demand. That is no small amount of power or energy and it is absolutely worth doing. It’s just nowhere near 100% of our electrical demand.

      • All of this will change when the cost of solar cells drops a good 50% lower than they are now in early 2017. When solar cells are much cheaper, it will then make economic sense to put panels on most southwest and southeast facing roof surfaces as well as the near-due south surfaces. That would most likely more than double the total panel area and increase total production to maybe 10% of our annual national consumption. Commercial PV plants could easily produce another 10%, totaling 20% of our electricity.

        Add in 30% wind, 40% nuclear, and 8% hydro and other, and that will leave only a few percent we’d need to get from standby natgas turbines.

        Good stuff.

Leave a Comment

User Name :
Password :
If you've no account register here first time
User Name :
User Email :
Password :

Login Now
Translate »