Speaking of Land-Intensive Solar Plants…


In remarks made at a business conference Friday, CEO Thomas F. Farrell II noted how much land solar panels consume. As paraphrased by the Richmond Times-Dispatch, he said:

A solar project that Dominion hopes to build near Culpeper would occupy 125 acres of land and power about 5,000 houses — about 30 percent of the time. By contrast, a natural gas plant in Warren County sits on 39 acres and can power 335,000 houses — about 80 percent of the time.

Writing in a Sunday op-ed pieceT-D columnist Bart Hinkle extrapolated from Farrell’s numbers: “Dominion can generate more than 24,000 megawatts of power all together. To get that from solar power alone would require more than 1,000 square miles of solar panels.”

Assuming Hinkle’s got his numbers right,  that’s roughly equivalent to putting Chesterfield County, Henrico County, the City of Richmond and half of Hanover County under solar panels in order to supply the state’s electricity needs. That’s a serious amount of real estate.

In all fairness, Hinkle’s example is somewhat extreme: No one is calling to convert 100% of Virginia’s electric power to solar. Indeed, Virginia’s Renewable Portfolio Standard sets a voluntary goal of only 15% for the state’s electric power to come from renewable sources, which also includes wind and biomass. And the state energy plan also calls for energy conservation as an alternative to building new generating capacity. But the example dramatizes how much land Virginia will have to dedicate to solar in order to make meaningful progress toward renewable energy goals.

It might be worth debating the pros and cons of converting thousands of acres of land to solar. We should discuss the environmental impact — what kind of wildlife would prosper in fields of solar panels? What would the visual impact be on habitat does fields of solar panels make? How do we minimize the adverse visual and environmental impacts?

I’m serious, solar panels could become one of Virginia’s most extensive land uses. This is something we need to start thinking about.

Update: Commenters have made the valid point that comparing the footprint of solar facilities to that of, say, a natural gas -fired generator is not fair unless it takes into account the right-of-way gas required for the pipeline that serves the gas plant, not to mention the gas storage and drilling facilities associated with the generator. Also, considerable solar capacity could be installed on rooftops.


Share this article


(comments below)


(comments below)


25 responses to “Speaking of Land-Intensive Solar Plants…”

  1. LarrytheG Avatar

    33 miles by 33 miles, right?

    take a look at southside, southwestern and Eastern Shore Va.

    or how about this – total up all the powerline and pipeline rights of ways… and add in the medians on many of our 4-lane roads.

    1. That’s a valid point. Assuming solar facilities are closer to the markets they serve, they require less transmission capacity — and less land for ROW.

      1. Cville Resident Avatar
        Cville Resident

        While true, that is still an enormous amount of land. Do you know anything about solar panels and topography? Do these solar fields require flat land? If so, such a development would be hard on SWVA. Most commercial and industrial development requires flat land. If solar fields require flat land, those localities would be giving up a very precious commodity in that part of the state for areas that are already very underserved in terms of economic development.

      2. Rowinguy Avatar

        All the transmission line ROW and distribution lines would have to be counted in the solar generation total also, since solar power doesn’t get from the panel in the field to the home in the suburb without these facilities.

        1. The transmission required was not counted for the traditional generating plants in the comparison so it should not be for solar. Besides if much of the solar is mounted on the roofs and property of commercial, industrial and government facilities no transmission would be required. Perhaps just some enhancements to the local distribution grid. This is a real potential savings for solar vs traditional generation that does not show up in any cost comparison of the various generating options.

  2. I posted this comment under the original article, but I thought it would be appropriate to repost it here.

    “Dominion also discounts solar because they say it uses more land than a gas plant. This highlights their prejudice for central station plants as opposed to distributed generation which has many economic benefits, including not requiring new transmission.

    A significant share of the solar contribution could be put on commercial and industrial rooftops or on their property without requiring additional land. This keeps energy production close to where it is used, improves reliability, resiliency, and potentially frequency control and volt/VAR benefits when coupled with some local storage. These are real economic benefits that can be bid into the PJM auctions by an aggregator to further lower costs.

    This only requires a bit of innovation and an appropriate mindset to lower costs and improve the customer experience.”

    There are also several additional considerations. Much research is being done to improve the efficiency of solar panels so that the same panel area would generate more electricity. Tracking mechanisms having also improved in reliability and have lower costs, so this also provides more output from the same area. Many parking lots would enjoy the shade provided by solar panels while they generate electricity for cars or business use. As more energy efficiency kicks in it will take far less generation to provide the same comfort that we require from huge plants today.

    Developers are also looking at allowing grazing under panels on sloping terrain. This is a bit more expensive, but does not take land out of production. Many new technologies are coming such as having windows and building surfaces be the solar generators. Dominion’s argument is a scare tactic to encourage the status quo rather than to encourage solutions that avoid the large land use.

    1. TomH, I agree, there’s a lot of scare-tactic involved in Tom Farrell’s comments, motivated by DVP’s fixation on the central-station generation model.

      Take rooftop solar, for example. I don’t know exactly what the percentage of horizontal surface devoted to rooftops is in a Virginia urban or semi-urban environment, but it’s likely that figure has been studied somewhere in considerable detail, in the context of surface drainage and runoff for Bay Act and water quality purposes. But let’s take a quick, crude look at it: look at Arlington County, VA in Google Maps-Satellite View and you would be hard pressed to say the amount of rooftop you look down on, relative to the total surface area, is less than 20%. Now discount that by 2/3 to allow for rooftops angled or oriented unfavorably and architectural obstacles like dormers; and discount that by half again to take account of structural obstacles to rooftop solar such as roofs shaded by trees or other houses, or roofs not owned by one occupant family (e.g. apartments not s.f. homes), and you’re still at 3.3%. So 3% sounds like a reasonable, even conservative, estimate of the potential land area available for roof-top solar by families so inclined in a denser suburban context.

      Now, 3% of 6,000 square miles (thanks to Dan Shaffer for that number) is 180 square miles. To get some perspective on what that means, consider that the original size of the District of Columbia (including what is now Arlington County VA) was 100 square miles (a square with sides 10 miles long). DC today is about 60 square miles. Taking Bart Hinkle’s numbers, if Dominion could (in the 30% of the time when the sun shines brightly enough) generate its entire capacity equivalent from 1000 square miles of solar panels, then 180 square miles is 18% of that. As discussed below, the maximum solar contribution generally considered feasible is around 15%. So, this amount could all be done on rooftops.

      This may give some sense of where distributed solar photovoltaic generation (DG solar) on rooftops could get us without the consumption of a single acre of farmland! There is absolutely no reason for the central station model to dominate our thinking, here. In some developed countries, like Germany and Israel, rooftop solar is already so prevalent that builders offer it as a matter of course — in some towns there are more roofs with it than without. Rooftop solar can be done, and is being done, enthusiastically, by willing builders and occupants, to the extent of more than 10% of total electricity consumption. And in addition, there are plenty of “barren” areas in Virginia where no current or potential farm use would be displaced by a large solar installation.

      I believe DG solar is in our future as well as Germany’s. But there is a natural ceiling, in the 15% distributed-generation range, that is hard to exceed cost-effectively and reliably; homeowner solar is not going to put Dominion Virginia Power out of business or render the grid irrelevant. What DVP has to do is avoid treating DG solar as an economic threat to its existence, but rather as an aspect of the future grid to embrace and build around. It’s a two-way street: The owner of an on-grid solar rooftop installation will necessarily know much more about electricity and the grid and the importance of secure energy supplies than the average consumer; treated properly, as potential political allies, these DVP customers could become DVP’s most important base of support in future decades.

  3. Speaking of on rooftops, I know those who do that. Either use the rooftops for solar energy, a greenhouse with veggies or both or otherwise you can $$$ persuade them to do so. I know people who have it on their rooftops. Seriously, what about the uninhabited places of Virginia or less inhabited?

    A good portion of my backyard is no chemicals and natural/organic/compost growing veggies and herbs. Between that and a few other things, I feed the house here, a couple of animal rescues, and the Foodbank. Last year it was over 100 pounds of food and its bigger now and will be by the time the season starts. Why waste land just cutting grass when no one will see it, when so many can benefit from a more effective use of it?

  4. Two words: Rooftop Solar

    It’s been said before, even in this thread. Solar on rooftops and in other under-used space (along the edges of highways have been suggested, for instance) is where the technology really shines. Of course, that’s not the kind of thing that a traditional electric company is entirely comfortable talking about. New business models and all that…

    Going back to the statement itself, it’s a bit rich comparing solar land use vs gas without looking at the land used by gas production, refining, and distribution. In one case, the costs are better internalized and therefore appear to be higher. Not to mention other external costs of gas, such as noise, water and air pollution.

    To be clear, this is not to say that solar is the best solution all the time or that gas generation is evil. But it also seems that this comparison was made by a person with plenty of motive to discredit solar. Flawed as it is, what purpose does a comparison like this serve aside from raising false alarm and spreading FUD?

    This man is threatened by solar. And if he refuses to adapt Dominion to emergent grid and generation technologies, he very well should be.

    Unfortunately, it’s going to be Virginia and Virginians that pay in the end. Either through higher costs, or through loss of corporate taxes when a more savvy out-of-state utility comes in (maybe from North Carolina, as with banking) and takes over.

  5. Dan Shaffer Avatar
    Dan Shaffer

    Looking at the National Land Cover Database, and selecting out only Developed (various intensities) and Barren land, one finds there is roughly 6000 square miles available in Virginia. Of course the majority of that won’t be suitable for solar, but if even 5% of it is…and I would expect that MORE would be…you’re already a significant portion of the way there.

    Yes, I realize that that is about 300 square miles, not 1,000, but it’s already roofs and parking lots…so there is no real “land consumption”. It just involves an upgrading of land use and perhaps a lowering of cooling costs and less of a heat island effect in urban areas. No better way to locate production near consumption than right on top or right beside!

    TomH and JPB know what I’m talking about!

  6. LarrytheG Avatar

    Mr. Farrell sounds not that different than someone who runs a taxi business in the face of Uber/Lfytt or the CEO of Kodak a few years back, or the current leaders of brick and mortar higher ed, etc

    He has a business model to defend -and he is demonstrating a failure to accept, to embrace changes – disruptive changes that will overwhelm the traditional utility business model -the same way that cell phones and computers, internet are disrupting other industries.

    The difference is that utilities have used their unique monopoly positions with govt to erect barriers to rebut the change and insulate against it.

    but solar is going to get built even if govt-enabled monopolies have hobbled it from achieving all it’s potential.

    And as, TomH points out – the grid IS going to DISTRIBUTE no matter DVP preference for central stations and it’s attitude that distributed solar is “unreliable”.

    essentially what we have is a 20th century grid in a 21st century world – and yes- IT IS going to cost BILLIONs to upgrade it – but blaming that on CPP is, in my view, a disservice -not only to Va rate-payers -but also to investors.

    no industry that positions itself in opposition to innovation – is going to prevail – the only part in doubt is the timeframe.

    1. “He has a business model to defend -and he is demonstrating a failure to accept, to embrace changes – disruptive changes that will overwhelm the traditional utility business model -the same way that cell phones and computers, internet are disrupting other industries.

      The difference is that utilities have used their unique monopoly positions with govt to erect barriers to rebut the change and insulate against it.”

      That is exactly how healthcare is.

  7. Peter Galuszka Avatar
    Peter Galuszka

    According to the Times Dispatch’s obedient stenography, Farrell also said that the utility’s four nuclear plants could fit in the Richmond Convention Center, which is utter nonsense, considering they had to build Lake Anna for coolant. Surry has the James River.

    I am continually amazed that these people get away with such statements and no one calls them on it.

    1. You make a good point here. The comparison is invalid. It would be like adding up the square footage of all the solar panels — without taking into account the space in between panels or the buffer areas around the panels.

      1. Nuclear plants also require a substantial exclusion zone surrounding the plants to protect the local population from potential leaks of radiation. The size of these zones vary based on the number of reactors on a site and the density of the local population. My recollection is that they typically contain at least several square miles of open land. Hard to get apples to apples comparisons from people with an agenda to push.

  8. Lots of ideas may sound good at first blush and may not be so practical to implement. This is not to rule them out in all cases.

    Use of highway medians, for example could require more expensive interconnection costs by virtue of their long narrow profile. In many areas the topography is not level, there are interfering trees and vegetation, and there would be the significant expense of Jersey walls or guard rails for protection against vehicle impact.

    Roof tops in some cases may be practical if flat for buildings not multi-story. A building with a 250′ x 250′ footprint if one story might have some possibilities for an economical installation. A similar footprint multi-floor building might have the same roof size but the proportional power production would be much less per floor.

    There is no escaping the fact that present technology requires significant area/watt.

    1. LarrytheG Avatar

      you have similar issues for all infrastructure – pipelines, powerlines, storm water, even highways are affected by terrain so you adapt to it.

      it’s not an obstacle than prevents any adoption.

      in terms of “interconnections” – how many bridges and other infrastructure like powerlines and pipelines already CROSS roads?

      You certainly would not slavishly try to put solar anywhere – but you WOULD put it where it make practical sense.

      my primary point here was to point out just how BOGUS the ‘ do you know how much land solar would require?” argument is…

      and it is…

  9. The same way the Governor can appoint MD’s to the State Medical Board and allow the Va. Dept. of Health professions to drop your civil rights and due process of law and only that person screams, rather than trying to get people together to say this is wrong and stop it.

    I could list a lot more.

  10. LarryG mentioned earlier DVP’s attitude that “distributed solar is “unreliable”.” I want to take issue with that ONLY insofar as “reliable” is a term of art in electric utility circles. In that sense, ‘reliable means ‘dispatchable at all times.’ Solar power is dispatchable only when the sun shines; and for practical purposes it is not dispatched then either because, when the sun does shine and the wind does blow, solar and wind are the cheapest power out there (they have zero fuel cost).

    Solar generation is highly ‘reliable’ in that it has a low outage rate for maintenance, etc. Solar is only “unreliable” in that the amount of sunshine is “unreliable” due to clouds and other weather effects and the daily and seasonal sun angle; other generation must pick up the slack when it isn’t there, and if the system has a high percentage of solar, the amount that must be replaced and the speed with which it must be replaced can become unmanageable. Imagine you’re the system operator and you have a large amount of solar generation in your control area and it’s a bright sunny day and there’s a major cold front rapidly approaching — when is all that solar power going to drop off? How suddenly? Do you have alternate resources running and ready to take the load?

    This is one of the reasons that solar power has a natural cost-effectiveness ceiling of around 15-20% of total generation. You can work around small amounts of solar given its fairly predictable normal daily cycle; but large amounts become increasingly difficult to back up, particularly in rapidly changing weather situations.

    1. Acbar is correct. This variability is a serious issue for utilities and the wholesale markets such as the one managed by PJM in our area. The large geographic range covered by PJM does allow for some evening out of the variability since on many days the intermittent cloud cover over solar facilities effects only local areas. But we all know that storm fronts can affect entire regions of the country and solar generation could be substantially reduced throughout most of the PJM region, perhaps for several days. Backup intermediate and peak load units must be available for those occasions.

      Lower cost storage might relieve some of this burden, but prices would have to come down a great deal for a large share of solar to be backed up in this way.

      I think demand response mechanisms under central control are likely to be cheaper ways of managing higher percentages of solar generation. In this way we will manage the demand to meet the supply rather than the other way around. There will likely be a long transition where many options are used to ensure reliability. The best way to make it happen is to have open and rigorous examination of the options such as PJM is attempting to do. I believe that this is more productive than Dominion declaring the options which pay them the best as our only choice. Dominion has talented people as does PJM. When given an opportunity to be innovative while still ensuring system reliability many good ideas might come to the fore.

      There are many responsible professionals in the energy industry. They just need the right environment in which to express themselves. They know what is at stake. They have committed their careers to providing reliable power. We just need to give them an opportunity to explore a wider variety of methods which might better serve the customer.

    2. LarrytheG Avatar

      re: ” ‘reliable means ‘dispatchable at all times.’”

      I think that is an artificial construct – that wrongly imposes restriction that makes no real sense.

      why in the world – would you use such an argument to not use an available resource?

      that’s sort of like the arguments of why you could not site solar on every site – therefore – you should not try to site it anywhere!!!

      that’s like saying that because you cannot site pump-storage reservoirs everywhere – that they should not be sited – anywhere.

      Solar is a harvestable resource – with limitations – LIKE ALL SUCH sources of power have. you cannot put Nukes anywhere… and you ought not to be planning them on a known earthquake fault (but that’s another story)… Coal emits highly poisonous mercury – but we don’t say it’s unacceptable to use it – we manage it.

      finally – coal and nuke is ALSO – NOT – immediately dispatchable… it’s takes hours to ramp it up – and the response to that has been – for decades – to run baseload plants – coal and nukes “hot” to keep them ready but turbines not turning until demand increases.

      so basically – we waste coal and nukes by running them when they are not needed – so that they CAN be immediately dispatchable when they are needed.

      we have gotten better at not doing that – as we have built more peaker plants – but when gas was too expensive – we ran baseload in a ready mode but not actually putting power into the grid – until demand ramped up.

      so we have somewhat similar issues – with solar… it’s got it’s own set of limitations that have to be worked around and accommodated so it can be used – efficiently – and reliably.

      that’s WHY I say that gas is the needed complementary fuel to run tandem with solar.

      but the truth is – if you don’t want to run baseload 24/7 – even when not needed – you also have to use gas as “peakers”.

      oh.. now tell me how I’ve got this wrong.

  11. Reed Fawell 3rd Avatar
    Reed Fawell 3rd

    My concern on highly inefficient use of land and view-shed has always been directed at wind power. What are the comparable numbers for wind farms?

    1. The actual ground footprint for large wind turbines is relatively small. Much smaller than a comparable 1 MW solar array for example. Which is why many farmers enjoy the steady lease income, while still maintaining most of their initial property use. Noise and bird impacts are still an issue though. As are the effects of the takeway transmission line that is usually required.

      Because wind speed is highest by the shoreline or at ridgetops, cropland is not usually affected.

  12. I like the idea of roof top solar, but I believe the EPA Clean Power Plan does not encourage roof-top solar (nor off-shore wind). The CPP focus more is utility scale solar (per the above picture) and on-shore wind.

    1. It is not clear to me either how the CPP will deal with customer installed generation. But I could see that it could be easy for the CPP to give credit for it, especially for larger installations on commercial buildings that might be installed by a few aggregators. If they were to bid this capacity in to PJM with associated storage, energy efficiency or demand response there would be specific contributions that could be tracked.

Leave a Reply