Green Roofs Could Help Improve Solar Panel Efficiency

There’s been a movement in architecture over the past couple of decades to help tie together large urban developments with plant life and greenery. We’ve seen a few buildings, and hundreds more renders, of tall skyscrapers and large buildings covered in vegetation.

The aesthetic is often a beautiful one, but the idea is done as much for its tangible benefits as for the sheer visual glory. Naturally, there’s the obvious boost from plants converting carbon dioxide into delicious, life-giving oxygen. However, greenery on the roofs of buildings could also help improve the output of solar installations, according to a recent study from Sydney, Australia.

The solar installation on top of Daramu House, Sydney, complete with greenery on the roof. Source: UTS Report

The study was led by Dr Peter Irga of the University of Technology Sydney, with a report published for the City of Sydney. The opportunity for the study came about precipitously, thanks to two similar office buildings located side by side in downtown Sydney. Each building had a photovoltaic solar system installed on the roof to generate electricity. On one building, plenty of plants were placed on the roof and around the solar panels, while the other building was left bare.

Over an eight month period, the roof loaded with greenery was 3.6% more efficient than the bare roof over the course of the experiment. The difference between the two was as much as 20% at peak generating times. This led to the green roof netting 69 MWh of electricity versus 59.5MWh for the bare roof. The extra 9.5 MWh generated over the period of the experiment is worth a full $2595 at local market rates.

The key to the difference in performance came down to temperature. Solar panels don’t work as well at higher temperatures, with Irga noting that “Temperatures above 25 degrees make photovoltaic panels less efficient.” This can be problematic in a place like Australia, where sunlight is abundant in the summer months but daily temperatures routinely span from 30 to 45 degrees Celsius.

A graph showing the difference in performance between the solar installations on the two buildings. Arrows A, B and C point to areas where urban shading caused a discrepancy.
However, note the large boost the green roof array receives at peak times. The cooling effect is most important during full sun conditions when temperatures are highest.
Source: UTS Report

Green roofs cool buildings through a process called evaportranspiration, or more accurately, the twin processes of evaporation and transpiration. Water from the soil and other rooftop surfaces is evaporated, reducing heat in the air. Additionally, small holes in the plants of the green roof, called stomata, are essentially the pores with which the plant exchanges gases with its surroundings. The plants lose water through these stomata to the atmosphere, further adding to the cooling process. Ideally, the vast majority of this water comes from rainfall, avoiding irrigation costs that can spoil the efficiency and environmental benefits of the roof as a whole.

Reportedly, temperatures were on the order of 20°C lower on the green roof compared to the otherwise identical bare-roofed office building. This is a remarkable figure, and one that speaks to the quality of the green roof design in the experiment. This comes down to careful selection of the right plant species, which are able to survive and thrive on the roof while also providing good cooling performance.

Thus, this significant temperature drop allowed the solar panels to remain in a much more efficient operating range, leading to that 3.6% efficiency gain. This figure was determined under simulated lighting conditions, in order to eliminate differences in the urban environment around the two buildings from spoiling the result. It may not sound like much, but huge amounts of money are spent every year researching for single-point percentage gains in solar panel efficiency. In comparison, providing a cheap natural cooling solution can have a remarkably outsized effect.

Garage with
Rooftop solar installations and green roofs alike must be designed carefully to deal with shading from surrounding buildings. This can impact the amount of power that can be generated, as well as require the use of appropriate plant species for the amount of sun available.

The green roof provides other benefits, too. The study reported that the roof absorbed almost 9 tonnes of greenhouse gases throughout the experiment, and reduced storm water outflows significantly as well. The plants were also much appreciated by the local wildlife. The team noted that insects and birds quickly flocked to the greenery. Even predator species were spotted up on top of the building, something that was surprising to see on a tower in the central business district of Sydney.

Overall, it’s a project that demonstrates a lot of net benefits. Additionally, it needn’t only be limited to green roofs. Other solar installations could benefit from co-located greenery which naturally cools its surroundings and leads to better solar array performance. Expect more research in this area, particularly in a localized fashion. Green roofs and similar technologies are highly dependent on local climatic conditions, and often need to be designed to work with the local flora and fauna as well. For those that dive in, it appears there are significant gains to be had!

83 thoughts on “Green Roofs Could Help Improve Solar Panel Efficiency

      1. Is that definitely not a translation problem? I don’t really see how the vertical forest would increase standing water for mosquitoes, but I can see how a lot of insects (not specifically mosquitoes) would enjoy it.

  1. Law of Unintended Effects.

    A client’s product line includes ‘green roofs”. This company discourages them for corporate customers, and refuses to do them for residential customers. They require much maintenance, special plumbing, and cost approximately 4x that of a conventional roof, and use more material, so get dinged where some local governments require construction to report on their ‘carbon footprint’.

    Another contractor that specializes in ‘green’ building projects said that their available green roof designs options will be severely decreased to reduce support calls.

    The best ‘green’ roof is one that is covered with solar panels.

    1. It’s more like the law of random unsubstantiated anecdotes, do you have any videos of cats climbing on solar panels? Or maybe a video on how to how to make delicious smoothies with a solar powered blender?

    2. it´s common where I live and it´s not more expensive than a traditional roof, but cannot apply to any type or building.
      Of course it´s not intended to grow trees, it´s just hardy green, not even foliage. No particular maintenance, quite the opposite: just some patience to let the plants establish in this 6 inches thick layer of gravel and sand

    3. Do you have more info about your claims? My building and several in the vicinity (I live in Hamburg, Germany) have green roof tops which don’t need much maintenance or increase our bills in any significant way. They are quite common nowadays.

      I think there is a lot of misconceptions about what a green rooftop is.

      1. 1. Altman Specialty Plants – was once a principal supplier of a hardy and unique hybrids, mostly succulents, designed to provide cover and to be water-efficient for use on ‘green roofs’ in the Southwest and Southern USA. Other than for government projects, they no longer offer this product line.

        2. Taiyo Yuden (USA) Inc. – used to offer products intended for control and monitoring of green roofs. The product line was dropped and the San Marcos division itself was later shutdown. The senior management was very risk-averse.

        3. Tamura Corp of America – used to offer distribution transformers for mounting in weird places on the building to support the various requirements of IEC, IEEE, and California energy efficiency, and ‘green building’ (LEEDS 2.0). They removed the spec for use on a ‘green roof’ several years past for reliability reasons and because of too many incorrect installations.

        4. Fluor Corp – major builder of corporate stuff and industrial sites. No longer offers ‘green roofs’. An engineer said that LEEDs 2 can make them a bit more difficult – no, I did not know why..

        5. Virta, Inc – builder of industrial processing plants. Now disallows green roofs (unknown reasons), but supposedly emphasizes energy efficiency.

        There are several more instances, but I have signed NDAs, so cannot say anything about their particular case.

        You people are getting a bit smug and too self-assured. After entry into my 6th decade, I am starting to understand whom should be ignored, what should be understood, and where the source of power lies.

        What I see here is innuendo and group-think taken to the point of religion.

        1. +10. There is no shortage of people making claims. There is a real shortage of people that are actually doing the work. I bought the green roof story and built one on my house. What a minefield it was. I am pretty sure we did it right but the cost and effort was wildly disproportionate to the return. And, I’m not even factoring in the amount time spent on research, trying to find contractors, ordering materials and designing parts that should be “off the shelf”. We wound up building it ourselves because no contractors would touch it.

          So yeah, the smug “greenies” need to get a reality check – I sure did.

        2. Seems like you are American and the post you are replying to is German… Perhaps over the pond you guys really don’t build ’em often (or well), but I’ve seen evidence of a growing number of them across Europe… Not enough data to really draw a conclusion on, but it seems like it might be very much a engineering standards type problem in the USA…

          Perhaps simply on grounds of cost – if this side of the pond building standards are higher, so the roof would cost a heap more and be nearly a green roof anyway to meet specs it is easier to deal with the tiny increase making it a green roof would entail perhaps..

        3. >”1. Altman Specialty Plants – was once a principal supplier of a hardy and unique hybrids, mostly succulents, designed to provide cover and to be water-efficient for use on ‘green roofs’ in the Southwest and Southern USA. Other than for government projects, they no longer offer this product line.”

          You mean the *SAME* Altman Specialty Plants whose current product catalog lists numerous green roof projects they’ve been involved in?

          https://altmanplants.com/page/18/?product_cat

          To this very day, this company specializes EXCLUSIVELY in succulents, which are a common choice in green roofs because they need little to no maintenance, and their excellent ability to retain water. You pretend that they’ve divested from this type of plant, when in FACT, it’s literally *ALL* they sell!

          Could you possibly *BE* a bigger SHILL?

          >”Taiyo Yuden (USA) Inc. – used to offer products intended for control and monitoring of green roofs.”

          #CITATION????

          >”The product line was dropped and the San Marcos division itself was later shutdown. The senior management was very risk-averse.”

          You are COMPLETELY full of CR@P!!! I trawled through ALL *TEN YEARS* of their website on archive.org (Wayback Machine) with the help of a little Python, and *NOWHERE* do they list any products “intended for control and monitoring of green roofs”.

          Taiyo Yuden makes inductors, capacitors, power supplies, and a number of GENERIC IoT modules. *NOWHERE* in their suggested applications appears the term “green roof”. The term has never once appeared on their website.

          Taiyo Yuden’s presence in San Marcos continues to this very day. Proof is literally a Google search away.

          >”Tamura Corp of America – used to offer distribution transformers for mounting in weird places on the building to support the various requirements of IEC, IEEE, and California energy efficiency, and ‘green building’ (LEEDS 2.0). They removed the spec for use on a ‘green roof’ several years past for reliability reasons and because of too many incorrect installations.”

          Another blatant LIE. Literally NOTHING in the Wayback Machine to support this claim. *ZERO* mentions of “green buildings” or “LEEDS”. Their current product lineup doesn’t even offer anything that remotely fits the applications you’re claiming.

          https://www.tamuracorp.com/products/electronic_components/category/trance/index.html

          What manufacturer on earth is going to change the specification for a product because an end user installed it wrong? You claim “too many incorrect installations”. Just how many of these “green roof transformers” were going bad? Given that green roofs are relatively rare to begin with, that number is in all likelihood *ZERO*. Why was there AC up there anyway? ALL solar panels produce *DC*, and any inverters are already going to be spec’d to output voltage that would directly interface to the building’s mains. What requirements of IEC, IEEE, and California energy efficiency standards would require obscure transformers in the first place. Clearly you have great knowledge of the situation, so you should have NO problem answering these questions.

          >”Fluor Corp – major builder of corporate stuff and industrial sites.”

          My god the utter Bullsh1t. The *very *FIRST* hit on Google for “Fluor Corp” and “green roof” is the steaming pile of BS you’ve been spewing here, right after the BOLD headline that says “It looks like there aren’t many great matches for your search”. In *NONE* of the remaining searches are those two terms directly related.

          >”Fluor Corp …. No longer offers ‘green roofs’. ”

          Maybe because they *NEVER* DID! The type of work they do has *NOTHING* to do with this kind of construction. They deal in huge industrial contracts making refineries and pipelines for oil and gas, mining and mineral extraction, and other industrial infrastructure applications.

          >”An engineer said that LEEDs 2 can make them a bit more difficult”

          “An engineer”. A nameless, faceless, likely MADE UP “engineer” that only YOU know about. Riiiiight.

          >”no, I did not know why..”

          More like you don’t know *ANYTHING* at all, except how spew lies in various online forums.

          >”Virta, Inc – builder of industrial processing plants. Now disallows green roofs (unknown reasons), but supposedly emphasizes energy efficiency.”

          Your WORST *LIE* yet. The term “Virta, Inc” “green roof” yields just *ONE* hit on Google. Your comment on HaD, and that is it.

          >”There are several more instances”

          No there aren’t.

          >”but I have signed NDAs”

          No you haven’t. You’re LYING out your rear.

          >”You people are getting a bit smug and too self-assured.”

          Wow. Awfully defensive for getting only TWO good faith questions.

          >”After entry into my 6th decade, I am starting to understand whom should be ignored”

          After 6 *SECONDS* of Googling your claims, it’s pretty clear that YOU should be ignored. You’re not even a good liar. EVERY claim you’ve made was trivial to debunk, and the more you went on, the more outlandish and weak your claims became. Did you *REALLY* think that readers here wouldn’t fact check that which stinks like so much dog dirt?

          Without exception, when articles about energy and climate change appear, follows a horde of ignorant, lying TROLLS out to denigrate practical solutions in order to maintain the status quo for their corporate masters. Or maybe you’re one of those cult members that hate empirical science, and reasoned, factual arguments.

          >”What I see here is innuendo and group-think taken to the point of religion.”

          I see it too, and it’s entirely embodied in YOUR posts. Better luck lying next time.

    4. It depends on the constraints. Do you need maximum number of panels? Or maximum output per panel. If you have room to spread them out so they don’t shadow each other, tracking panels have a 3-fold increase in output over the day. I wonder if there is some Python code out there that finds a linear programming solution given the geometry, costs, power needs, and a handful of other constraints.

      An entire array that rotates like a Lazy Susan is ideal

    5. “This company discourages them for corporate customers, and refuses to do them for residential customers.”

      That pretty much leaves government buildings, but not to worry, governments can “afford” the added expense.
      B^)

    6. This is true. I built a small green roof area (200 sqft) on my house. What a PITA to deal with. We wound up doing it ourselves (2 contractors bailed on us), had to add plumbing to the roof, had to build the planting trays (no one would sell them to us) and had to special order the liner materials at significant markups. Also, it is a high maintenance, high cost part of the house. Took significantly more labor (3-4X) than a traditional roof. Honestly, I would never do it again.

      And, yes, the specialized liner adds to the carbon footprint significantly.

      This was 2 years ago. One of the things that really bugs me is there were all sorts of people extolling the virtues of green roofs. Where were they when it came time to actually build one??? This emperor has no clothes.

        1. Depending on the plants and the climate zone you are located in, you don’t only have to plumb to get water off the roof, but also a line of water to irrigate the roof plants (if you do not have succulents and want to have the cooling effect in the summer, you have to put somehow water on the roof for the plants to evaporate it).

          a tin or shingle roof is the same all over the globe.
          a green, planted roof is different everywhere.
          Some are pretty low-maintenance and sturdy like the ones covered with birch bark and birch tar and grass sod on top of it. low-tech, local materials in scandinavia, and over 80 years old when I visited 25 years ago. That said, NOT entirely maintenance free! someone has to rip out the tree seedlings each year, because the roots would go through the roof somehow, and if you rip them out or cut them after the roots protruded the roof, then you have water leaking in and the roof rafters will foul and fail later.

          A supermarket close to my office had a green roof with earthenly reddish brown puffed clay substrate and hardy mediterraneran herbs. It was cool to sit in the roof terrace of the restaurant there and see the bees and the birds living on a roof which had lavender, thyme, and other such well scenting herbs on it.

          They have been sold following the 2008/2009 crisis, and since then I can watch the roof slowly fail.
          Thyme and heather has taken over, 4-5 years old birch seedlings piercing the sealing layer.

          so: green roof is good, provided that once a year someone can make corrections.

          Also you want to have a very simple roof geometry with almost no protrusions, because that is where the sealing will fail eventually, in all the nooks and crannies.

          1. This seems to be missing a bit of context for the conclusions to make sense. If by “you do not have succulents”, you mean that if you choose not to put succulents then you need plumbing to water the plants, the obvious question is why then you wouldn’t just always choose succulents.

            The plumbing down and back up seems overly complicated too, so I’m not sure what I’m missing. Surely the soil or other layers are intended to absorb some rainfall, so I’m not clear on why you would ever plumb water away, assuming you mean something more than an eaves trough or similar for occasional overflow.

            For irrigation, presumably you’d choose plants suited to local rainfall patterns, which you implied since these green roofs are different everywhere, so I’m not sure what the irrigation would be for. Unless the roof was simply not designed according to the local climate, and was instead designed for other reasons, like a garden used by a restaurant below.

    7. >”This company discourages them for corporate customers, and refuses to do them for residential customers.”

      Just who does that leave left exactly? If not corporate/commercial, and not residential, then there’s really not any meaningful business left, now is there?

      >”They require much maintenance”

      All these points just *REEK* of utter BS. I can’t find a *single* credible source that recommends ANYTHING that isn’t specifically chosen to be low maintenance. Low maintenance is literally the FIRST criteria the majority of sites about green roofs mention.

      >”special plumbing”

      Utter BS. Literally *EVERY* flat roof in existence already has multiple roof drains, which don’t even require any modification to accommodate a green roof. This claim is *ENTIRELY* made up. Feel free to cite something credible to support this nonsense.

      Also, green roofs play a *significant* role in ameliorating storm runoff from overloading city storm drains. Almost without exception, every source I’ve posted up and down this thread mentions this.

      >”and cost approximately 4x that of a conventional roof”

      More FUD.

      The very first finding of this gsa.gov cost/benefit report on green roofs:

      “Compared to a black roof, a 3-inch to 6-inch green roof covering 10,000 feet has a Net Present Value of $2.70 per
      square foot per year, Payback of 6.2 years and an Internal Rate of Return of 5.2% nationally”

      https://www.gsa.gov/cdnstatic/Cost_Benefit_Analysis.pdf

      This empirically based report alone obliterates *every* unsupported anecdotal BS claim you’ve made.

      Here’s another source that totally counters your fallacious arguments:

      “The university found that the green roof would cost $464,000 to install versus $335,000 for a conventional roof in 2006 dollars. Over its lifetime, though, the green roof would save about $200,000. Nearly two-thirds of the savings come from reduced energy needs for the building below it.”

      https://www.thestreet.com/personal-finance/mortgages/what-a-green-roof-costs-you-on-the-way-to-saving-everything-13161050

      >”and use more material, so get dinged where some local governments require construction to report on their ‘carbon footprint’.”

      FUD, FUD, and more FUD. Care to explain why so many local governments are pushing for solutions like these?

      https://stormwater.pca.state.mn.us/index.php/Cost-benefit_considerations_for_green_roofs

      https://e360.yale.edu/features/green_roofs_are_starting_to_sprout_in_american_cities

      https://www.nytimes.com/2019/10/09/realestate/the-green-roof-revolution.html

      https://www.syracuse.com/business/2021/07/how-7-acres-of-grass-grown-in-syracuse-area-ended-up-covering-nycs-javits-centers-roof.html

      >”Another contractor that specializes in ‘green’ building projects said that their available green roof designs options will be severely decreased to reduce support calls.”

      “Another contractor” STINKS every bit as bad as tRump’s nebulous “many people” that end up being the ignorant voices in his head. I’d bet any sum of money that you can’t cite a credible source for this claim. It’s ENTIRELY devoid of ANY specific factual information. It’s just another of your crappy, completely unsubstantiated anecdotes.

      You’re either woefully ignorant, or intentionally shilling for someone. NO ONE is this clueless by accident.

      1. Instead of mowing, keep goats on the roof as well :) The Goats can provide milk, cheese and meat!

        On a more serious note, It would be interesting to see the ecological and economic cost of spraying though, you’d need to use purified / filtered water so that is an extra cost, plus the power consumption of spraying it, but spraying directly onto the face of the panel would evaporate a lot more heat from the panel.

        I wonder if you could use some kind of water cooling and use the heated water in the buildings heating system. Again, you’ve got to add the cost of running the pumps / maintenance etc but it would be an interesting study and you could probably lower the temp by more than the grass.

    1. Highly reflective paint is a dumb idea. In most place of the world, you spend more energy heating the building than cooling it. You’re not going to paint your building twice a year, so if you use such reflective paint, you’ll end up with an actual increase in your heating bill compared to without it. Plants are convenient here, since, in winter, they provide additional insulation while still capturing the few rays that hit your building, and in the summer, they increase the humidity of the surrounding air, cooling the building (and the solar panel if any).

      1. Highly reflective paint is an excellent idea.
        Highly absorptive paint is a dumb idea. In summer, the sun is high and it’s hot, roof temperatures get very high, and your air conditioning burden increases dramatically.

        In winter, the sun is low, so there’s not much thermal gain through the day, but there is enormous radiative loss during the long night, increasing your heating burden.

        And the thermal cycling of a dark roof is much greater than a white one, decreasing its longevity.

        Dark roofs are a dumb idea.

        1. For temperate zones, use a white roof with long eaves, and black walls reflecting the high sun of summer and absorbing the low hanging sun of winter!

          (I once proposed venetian blinds that were reflective on one side and flat black on the other, to reflect the sun in the summer and to absorb it in the winter, but the wife didn’t go for that idea…)

          1. >”use a white roof with long eaves, and black walls reflecting the high sun of summer and absorbing the low hanging sun of winter”

            If this one simple design requirement were made mandatory on all new construction and retrofits alike, we’d see an enormous downward shift in overall energy consumption. It amazes me that it’s not even recommended by the various agencies recommending climate change mitigation strategies. Probably the single most cost effective thing we can do.

        2. Paints like titanium white have high reflectivity in the visible and high emisivity in the IR, which is why they are used on observatory domes. They are surprising low temperature under noon sun and cool quickly at night. Just use good roof insulation.

      2. “Highly reflective paint is a dumb idea” – if you leave in a cold climate somewhere.

        Here in Aus it is a really good idea – note that our winter temperatures can be about the same as your summer ones….
        I took my holiday houses old roof (very dark steel done in the 70’s) and replaced it with the lightest one I could. Also repainted the house from dark color to a light one.

        The differences was enormous!

        On solar panels, I’m surprised more work isn’t done on other means of cooling, and if they are energy positive, ie air via fans, heat sinks, water etc..

        You have to do something when things left in the sun can go to 75C plus fairly quickly…

        What is happening the most around me is that people are putting most of their panels on the east and west side of their roof – with the minority to the north. Thus they get more power when it is cooler in the morning and evening.

        1. If Aus follows European daily power use curves that is also the times of highest domestic demand – so having more panels optimised for the morning and evening often makes sense – you won’t hit the same peak power and probably still a lower average compared to the same amount of panels pointed to the noon day sun (as those panels still work somewhat effectively through morning and evening, where while the sun is in the east the west facing panels are basically useless and visa-versa). But there is no point making lots of excess electric at lunch while nobody is home unless you have some national industry that can soak up all the power you can feed it (and as the homeowner you get paid for the exported power enough to be worth it)…

          I wonder if active cooling of a solar panel does make much sense, its more to go wrong, adds construction and maintenance complexities, so might be better to just have more simple panels… That said something simple like giving them a good Al back side heatsink to radiate the extra energy more effectively could be done, and doesn’t add much complexity (though perhaps too much capacitance)…

      3. >”Highly reflective paint is a dumb idea.”

        Literally EVERY credible source completely disagrees with this claim.

        https://www.energy.gov/energysaver/cool-roofs

        https://heatisland.lbl.gov/coolscience/cool-roofs

        https://www.epa.gov/heatislands/using-cool-roofs-reduce-heat-islands

        https://www.osti.gov/servlets/purl/791840

        >”In most place of the world, you spend more energy heating the building than cooling it.”

        Key weasel word: “spend”.

        For residential, it takes FOUR TIMES as much *ENERGY* to heat a home than it does to cool it:

        https://www.bloomberg.com/news/articles/2019-07-10/why-we-always-fight-over-air-conditioning

        In a study of US office buildings, AC accounted for 14% of energy consumption, while heating accounted for only 5%:

        https://seor.vse.gmu.edu/~klaskey/Capstone/MSSEORProjectsSpring18/SUCBA/downloads/SUCBA_FinalReport.pdf

        Per BTU, air conditioning is wildly MORE expensive, but typically the change in temperature required by the user is substantially smaller for cooling than for heat, which is where the cost savings are. At any rate, the use of air conditioning is dramatically on the rise.

        https://www.theguardian.com/environment/2015/oct/26/cold-economy-cop21-global-warming-carbon-emissions

        >”You’re not going to paint your building twice a year”

        No, that’s a ridiculous straw man argument, and doesn’t actually support your fallacious claim that using a reflective pain on your roof is “dumb”.

        >” if you use such reflective paint, you’ll end up with an actual increase in your heating bill compared to without it.”

        Pure FUD: https://en.wikipedia.org/wiki/Fear,_uncertainty,_and_doubt

        In the summer, when the sun is much closer to being overhead, FAR more energy is absorbed by rooftops than in winter when the sun’s angle is low. The energy saved in AC costs will by FAR out strip any additional expense in heating from lost heat from the sun in winter, especially when coupled with better insulation and energy efficient windows.

        1. Your answer is very detailled.

          In my home, the number you are presenting don’t work out. Why ? Because I’m heating my home with wood, but use electricity for cooling. Heating requires energy for 10 to 20°C while cooling is only for 2 to 5°C. Sure, wood is cheaper, but it’s not 4x cheaper.

          Also, about the sun position and its impact, I’m measuring it and it’s not like your are writing. Yes, the sun is higher in summer (in my location it’s 70° in summer and 24° in winter), but the roof is oriented smartly for it at 45° or so (depends on your latitude). It’s basic math going on here: in the winter, you’ll have cos(24°) reduction of the optimal absorption (that’s a factor of 0.91, not that bad), and cos(35°) ~= 0.81 increase in the summer. All in all, it’s too insignificant to make a difference with the 4x increase in energy required.

          Obviously, insulation only makes things better and it’s clearly the way to go first by a large margin, but it’s not the only way to improve the situation.
          Clearly, having the right ventilation reduce the need for AC cooling “most of the time”, and I’m speaking of canadian well for example and a correctly triggered fan (at night for the summer and during the day in the winter). The size of the building and its shape is also very important, American home are the example of what you shouldn’t do, a good building should be as compact as possible (prefer a 2 floor rectangle to the same area on a single floor).
          The material used in the building is also important, because it’s able to regulate the humidity and increase the “feel good” zone of the building (if the wall perspirate, you can cool your home with an evaporator without needing the AC’s consumption).

          As a matter of judgement, I’m more likely to believe on what olders did than the new hype miraculous paint. Sure, in Alger the home are white, but they need to cool even in winter. In 45° latitude, the old home are dark because well, ancient were not stupid, they realized it was better to heat them than the white version.

          1. Sweethack: in winter you will lose more heat with a dark roof and walls than you will with white ones. There is an ENORMOUS amount of heat radiated away.

            And since in winter you’re in the dark more than in the light, you’ll lose more to the dark than you will gain in the light. With a wall or roof temperature of 0 C a dark surface will radiate away 200-300 watts per square meter. A light surface could be a quarter of that.

            How much of your 45 degree roof is absorbing sunlight? At most half of it. How much is radiating away heat? ALL of it. As are all your walls too.

    2. >I wonder if simply painting the roofs with highly reflective paint

      The short answer is “yes!”

      “Paint the roofs white” was suggested over a decade ago to combine “making jobs” with “reducing energy use.” Here’s a brief piece in The Atlantic from 2011: https://www.theatlantic.com/business/archive/2011/07/paint-your-roofs-white/241784/

      Oh and if you want to see the debate that went on at the time, here’s the relevant conversation on Slashdot: https://politics.slashdot.org/story/11/07/18/1951240/bill-clinton-says-paint-your-roofs-white

      Now that more roofs have solar panels on them (compared to 2011) the idea seems to make even more sense, for the thermal reasons in this Hackaday article.

      What I haven’t been able to find is whether any permanent change has come from the proposal. Have any building codes been changed to encourage light-coloured roofs?

    1. That would just reflect back onto the underside of the panels and make them hotter. This cools the area because of the water used by the plants. I bet if you just used the same water you use on the plants and sprayed the panels, you’d get them even cooler (corrosion issues aside).

      1. “You remove” _a large amount of_ ” the covering,” _hopefully find where the leak is (the actual leak in the membrane could be 10 meters from the drip point in the ceiling)_, fix the leak, and then replace the covering”,
        _hopefully not causing new leaks in the process_.

        FTFY

    1. Well, most green roofs are made with modular planting trays that can be removed. Best practices use multiple layers of film/membrane type products so a leak is pretty unlikely. The possibility of a leak is not a significant drawback – there are plenty of other reasons green roofs are a bad idea., though.

    1. >”Probably cheaper and less maintenance to add cooling fans to the solar panels.”

      Key weasel word: “Probably”

      I’d like to see an explanation how hundreds of additional power *CONSUMING*, MECHANICAL points of *FAILURE* could ever be “cheaper” and be “less maintenance”, compared to a *plant*, that takes ZERO electrical or fossil fuel derived energy, has *ZERO* moving parts, and is capable of self-healing and reproducing without ANY outside help.

      Do people even think critically anymore?

      1. Plants in such a situation need tending, and that currently requires a human gardener to go round keeping them under control.. SO it is probably cheaper on both, sure its mechanical parts that can fail, but its a cheap fix to do so, takes seconds, so the one guy employed can work on potentially hundreds of sites, where the gardener probably has to spend all day at each site a few times a month, so can’t do half as many sites.

  2. It’s common knowledge that cooler panels give higher output. Their voltage increases and that results in higher output.

    What many don’t realize is that solar panels generate less heat the more power that is drawn out of them. The process of converting photons into electricity means there are less photons to heat the solar panel. Thus the higher the efficiency of the panel, the less warming it potentially produce in bright sun light. Perhaps another way to reduce global warming is to convert those photons into electrical power instead of heat by installing lots of efficient solar panels.

    These effects were very evident with the solar power model airplanes I built. In the late 90s I built several models that set FAI world records in the solar power, no buffer battery class. When first launched their performance was marginal but as the cells cooled due to air flowing over them, and their reduced self heating the plane’s performance increased greatly.

  3. WoW Imagine solar panels covering water canals in the desert! Shading from the solar panels reduces water loss to evaporation, and cooling from the canal will boost the solar panels. Imagine both water and electric companies benefit. Synergy is wonderful!

  4. If you want to know the effect of temperature on solar cells, then just look at the bell curve of any solar installation during a partly cloudy day. There are “gaps” in the “bell” because of the clouds (and the panel cools down), and each time the sun falls on the panels again there is a short peak that is significantly above the bell-curve, until the panel gets hot again.

  5. While this story certainly hits all the markers for clickbait virality, I am highly skeptical of their result. Did they never water their plants? If they did, did they factor that into the energy costs? What about the added wear and tear to the roof caused by soil/water, cost of the plants, soil, and maintenance? If it truly did lead the a higher return on investment, then it is a remarkable story, but I think it more likely that these ancillary costs were overlooked.

    1. How much higher would be acceptable to you?
      Commercially-available cells (21-25%) are already most of the way to maximum theoretical efficiency (30-33%) for single-junction cells. Your average rooftop panel at 15% is not too far behind.

      Exotic and expensive multi-junction cells can double both those numbers, and it’s unlikely to ever get much better than that now.

      If you want to get extra work out of your sunbeam without breaking the bank, concentrate the light, filter it so only the “useful” (red & infrared) photons hit the cells, and use the rest for thermal loads, Stirling engine, or growing plants.

      A layer of water with a red dye makes a fine thermal absorber, letting through only the efficient red photons.

      A “200 watt” panel produces 200 watts with a kilowatt of white sunlight falling on it.
      Hit that “200 watt” panel with a kilowatt of red light and you’ll get 400 watts out of it.
      Now, it might take 10 kilowatts of sunlight to get that 1 kW of red light, but now you have 9 kW of concentrated blue, green and infrared light to do whatever you want with. And cooler, more efficient solar panels too.

      1. Is 15%, or even 30% efficiency good? Seems a low bar to set for something. Certainly not good for the internal combustion. I’m not entirely sure what the effeciency value is for coal, gas, hydro, wind or nuclear.
        I have read articles over the years on this or that technology showing good signs of improving effeciency of solar power through either improving existing processes or sometimes new materials, and the like. All exciting stuff. I recall an article about solar ‘shingles’ which I thought was pretty good as this has more than just solar power attatched to it. These articles have been flowing for a really long time. Yes improvements, what I would expect from how technology presents itself, no. Seems we aren’t all that much farther along than from a popular science magizine from the 80’s with solar power. Batteries, yes by quite a bit, controllers and all the electronics involved, wow, huge steps. Not so with the panels. I would like solar to be in a place where I don’t get door to door salesman coming to my house selling solar because there is no need for that from a consumer standpoint. I don’t get door to door car salespersons, or appliance salespersons for that reason. If solar remains low effeciency (whichever technology it is), or is termed ‘high effeciency’ with number like 30%, uphill battle. I guess I could just be expecting more out of technology, we are flying a helicopter on Mars after all.

        1. Most Internal combustion engines are at around that top efficiency compared to maximum potential in the fuel too – the difference is sunlight is freely available, petrol etc has lots of extra extraction, refinement and shipping costs… So in terms of practical efficiency a solar panel could be be considered basically infinitely efficient – it takes something that would exist anyway and makes it useful at nearly no cost to use…

          As always you have to look at what you actually mean – a solar cell turns perhaps 20% of energy falling on it to electric, so it sounds poor, but in real world terms its practically magic free energy as the sun isn’t going anywhere anytime soon, and can’t be throttled down so that energy was going to be put to no use otherwise… So your solar household might be getting all of its energy requirements or more likely something over half for nearly no cost – the cost involved being cleaning them ‘often’, replacing batteries (if you have any – which isn’t required really) and perhaps the MPPT inverter box every 10 years or so (but the panels themselves should last basically forever with some performance loss).

          If you could capture more that is great, as that means a smaller area and less resources are needed to provide the same power, but when you are talking about fit and nearly forget about practically infinite power sources its not the end of the world to be stuck at 20% of the theoretical maximum forever…

          Also there has been a major improvement over the last few decades in Solar, power output degradation – the panels from the early days still work assuming they are undamaged, but they dropped from their peak to possibly even less than half the original output pretty fast, perhaps a few years, now panels pretty much don’t degrade measurably so should stay really really useful power generation devices for decades…

          1. Unfortunately I have to wait for someone smarter than I to improve or invent something new in the solar department, as I’m stuck as a user due to my natural limitations. I am positive that I am glossing over many technical details. I’m stuck wondering often about the solar industry, fully aware that materials have limitations. I did a very quick search for solar history timeline on the net, not that everything is true that I read, but I have a slight bit of confidence if properly crossed referenced with sources offering general information not related to specific queries. I see in 1999 (approx.) a company achieved something like 31% efficiency in their trials through a process different from what was done before in manufacture of panels. This makes me wonder why I’m still seeing lower numbers, in some cases a fair amount lower, realizing a few percent in terms of solar is a pretty big deal. It is a fair assumption something didn’t work out, but it was achieved, which is a valuable thing to just do in itself. If 32.something % (or there abouts) is the very best we can do in todays technology, which doesn’t seem toooo far off from what was *possible* 20+ years ago, I have to wonder if tech is looking into to newer materials hard enough, or if band aides are the real order of the day. Honestly, what I’m left wondering many times when I read acticles of today, compared to acticles of yesterday is how serious are we (collectively) really about this. I enjoyed reading this article, thinking about grass on roofs, which is also a very, very old concept, same time I’m disappointed. Can’t stop thinking band aide.

          2. @Chris you have to remember its a very different thing making something that works once, or for a week, to something that lasts – modern solar panels of the common sort are now at the sort of efficiency range you could just about manage to get for 5 mins in 1990 something and they stay there outputting at that sort of level, probably for decades, certainly for at least one – the degradation issue used to be really bad, rapidly trending down to something like half the initial performance, now the down trend is there, but its soo small it is almost irrelevant, and seems like it will ‘stabilise’ at a much higher proportion of the original output (though with such slow decline we are making the assumption it will follow the same degradation curve just at a slower rate – maybe it can still get down to the really low proportion of its start, but if takes decades (and it definitely will) that is a massive win)…

            Also got to point out with better coatings and construction methods the panels can ‘self-clean’ pretty effectively, work more effectively with off axis light, in lower light levels – so in practical real work use they are miles and miles better.

            In Short there have been great strides made in solar tech, just because the often very short lived, in some cases bio derived panel techs can do better at one or other key statistic, like output doesn’t mean the stuff being commonly fitted and used isn’t massively better than 20+ years ago, it just doesn’t win across the board in top trumps style stat making… (and won’t be the right type of solar panel for every job you want solar energy for)

      2. That really depends on the type of solar panel OP has. Every technology has a different absorption spectra, as can be seen in this image:

        https://www.researchgate.net/profile/Chris-Gueymard/post/Which-part-of-solar-spectrum-is-more-useful-in-PV-technology/attachment/61423981d248c650eda44cb0/AS%3A1068367414833152%401631730049462/download/Spectral+response+compar+v2.png

        Wavelengths shorter than the red end of the spectrum *DO* contribute substantially to the overall output, and filtering out all but red and IR is guaranteed to considerably reduce their total output, while doing almost NOTHING to cool them, given that IR is the greatest contributor to heating them in the first place.

        1. And a perfect example of “a little bit of knowledge is a dangerous thing”, or “how to lie with graphs and statistics”.

          A solar cell will absorb a photon when the photon energy is greater than a certain threshold. For silicon, it’s around 1.1 eV, or 1 micron wavelength. That one absorbed photon will produce one electron worth of current, at around 0.5 volts (if it doesn’t get squandered by other processes first). So, an efficiency a bit less than 50% is the very best you can do with a single material.

          The important part: any excess photon energy will NOT produce more current (or voltage), but simply gets wasted as heat. So if a blue photon hits the panel, it still produces just ONE electron output current, still at 0.5 V, even though it deposited 2.5 eV on energy into the silicon. So, the BEST you can do (with silicon) is get 20% efficiency out of blue light.

          Yes, bluer photons get absorbed. Yes, they produce current. But unless the material is matched to the input spectrum, you end of throwing away most of the photon energy.

  6. I don’t really understand why they’re using materials with high heat soak ie aluminium, galvanised steel, when there are alternatives? Roof construction eg corrugated iron, hasn’t really changed in a couple hundred years and wasn’t designed for PV systems. Surely this has been covered? Especially if they’re doing anything for single point gains.

  7. Why not combine photovoltaic panels and solar water heating to take the heat away during the higher temperatures? While I like the idea of using vegetation on rooftops, if the goal is to remove heat you can control the amount of heat removed by varying the amount of water being heated. That would reduce the power required to heat the building as well.

    1. How hot would you want that water? 50C? How much does a panel lose in efficiency at that elevated temperature? Will it be worth the effort over, say, just having a separate dedicated (and more efficient) water heating collector panel?

      1. Of course you want the water near boiling temperature, or at least above 70c to inhibit bacteria growth while at the same time keeping the solar panels at sub zero.

        Or:
        It’s not so easy to do something useful with small temperature differences.

        1. The panels don’t need to be sub zero, and you don’t need to worry about bacteria growth – use a heat exchanger with pure alcohol for example in the panel side – then it doesn’t matter what the sustained temperature on the panels is. And you can dump all that heat into the ever cycling ‘hot’ water system (which limits bacteria grow potential – as its only warm but not boiled for short periods), saving you energy on your hot water as its not heating mains cold water any more..

  8. A lot of these have been called out already. I have dealt with green roof projects on several buildings.

    1. Drawing birds and bugs to the area you have solar panels is not normally something you want. They get the panels dirty and make a mess of things.

    2. The building structure has to be significantly upgraded for weight and water handling. This means more materials are used so you have to calculate the “greenness” of the offset. There is a very significant maintenance workload which also has an impact on overall “greenness”. Is the area covered with plants actually offsetting the truck rolls, watering, and maintenance required to maintain it?

    3. If you have enough area for plants around your solar installation maybe you would be better off with more solar panels making use of that area.

    4. The carbon capture of the plants on the roof is very temporary, as they die or get cut back the carbon contained in them is released back. If the plants grow rapidly (catching lots of carbon) you have to prune them. If they grow slowly requiring less maintenance they are catching less carbon. Trees can store carbon for hundreds of years but grass and other shorter lived plants store it until it is cut and decomposes. Remember that the carbon stored is a function of mass.

    5. If you are in an area that gets significant snowfall, a roof with vegetation make snow control around your panels very complex. It is easy to put in melting systems or use manual clearing methods when the roof is flat, if you have vegetation it is not so straight forward and removing the snow cover from the vegetation also exposes it to cold weather damage. For some of us in the northern regions, a green roof is only green for about half the year.

    6. I think the temp difference is going to be highly site dependent and only will be significant for certain weather conditions and times of the year. You would have to calculate how much of the time there is a positive effect to determine the cost/benefit ratio.

  9. That $2595 will easily be offset by the cost of gardener and maintenance. Not to mention the added weight. In Hong Kong, a “green roof” like that in a university collapsed after heavy rain because the soil soaked up all the water. Why not just add cooling then? Because “green roof saves energy” sounds like a better topic for PhD paper?

Leave a Reply to sweethackCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.