Harvest the sun twice: Agrivoltaics promises sustainable food, energy and water
109 comments
·February 27, 2025hannob
0_____0
Great points
Some (very rough) numbers to give scale here:
Corn in the US covered ~90 million acres in 2024. Something like 40% of that gets turned into ethanol IIRC.
Based on solar capacity figures around ~200GW, the surface area of the PV panels themselves (not the installation) are about 250-300k acres.
Both solar and bio-ethanol render an energy output on the order of 250-350TWh/year. Broadly similar levels of energy, but using a tiny fraction of the land area and chemicals and industrial pollution for the former.
daveguy
I think this implementation and purpose is slightly different. The climate they are testing in is East Africa. One harvesting is the photovoltaics for energy. But the second harvesting is for food crops. The partial shade provided by the PV panels allows the plants to lose less water from evaporation and protect against overexposure to sun so that food crops like beans and corn can provide higher yield in a normally difficult climate zone. I agree harvesting the sun as PV + biofuels is a step down, but PV + food crops in a difficult climate zone is a step up.
nanomonkey
Don't throw biofuels out, there are plenty of fuels you can make from the parts of plants that we do not eat. Corn stover, walnut shells, bigasse, olive pits, etc. are all unused components that can be converted into fuel sources through gasification, anaerobic digestion, fermentation, etc. into usable fuel. Many times it's better than letting these leftovers rot in the fields and release methane into the atmosphere.
Honestly, I'd like to see some of the excess solar energy put into up converting such waste streams into storable fuels that can be used as denser alternatives to batteries when needed (winter heating, air travel, etc.).
gamblor956
Also, biofuels have >10x the energy per unit of mass compared compared to even the best batteries in the lab (and some sources put that at closer to 100x). There are a lot of use cases for biofuels (as a replacement to petroleum) that batteries can't come close to handling.
Converting the waste stream from agrivoltaics into biofuels appears to be the best option: reduced land for "energy" crops, increased output for food crops, and biofuel to replace petroleum.
542354234235
I feel like hydrogen is the answer to fossil fuels, not biofuels. Battery electric vehicles (BEV) are great for personal vehicles. But for heavy transport, batteries just don’t work. They are too heavy and can’t be “refueled” quickly for continuous operations. We are already having issues with having too much solar power and not enough demand. We could be load shedding that by using it to create hydrogen.
We have gas stations everywhere because individuals need that kind of convenience to operate comfortably. BEVs have the same infrastructure in that every outlet is a charging station. While the barrier to entry for hydrogen is too high for personal vehicles, the logistic hub nature of heavy transport would be much easier to convert, using hydrogen storage/filling stations at strategic location to support hydrogen powered semi-trucks, local delivery trucks, busses, etc.
corysama
Converting waste to fuel is great.
Meanwhile, I’m pretty excited about this project to make it cheap and easy to convert solar power to liquid fuel. Starting with carbon-neutral natural gas.
salynchnew
Correct arguments all around re: utility scale solar, but the incentives for an individual farmer to immediately adopt are huge: local generation makes power cheap, shading makes water use more efficient, etc. Obviously there is a capital cost for the system and costs for maintenance, etc., but given the trend of increasing global energy costs and the resulting burden on households (https://www.nature.com/articles/s41560-023-01209-8), it seems like a very nice idea.
rendaw
The article's arguments don't revolve around landuse, and in fact the article doesn't discuss landuse at all.
explodes
In rural parts of Colorado, I have seen large signs on farm building protesting the use of agriculture land for solar.
NathanKP
The main advantage that I see in agrivoltaics is for the purpose of density. If you can get both power and food production in a more dense footprint, located closer to a population center, then that will be much more efficient for purposes of transportation. Both electricity transmission and food transportation get worse across larger distances.
ianburrell
Power transmission is really efficient, HVDC is around 3% per 1000km, AC is around 6%. It is more efficient to put solar panels in sunny areas and transmit across country than have less efficient local ones. Oregon where I live is bad for solar with lots of rain, it is better to put panels in California desert and bring up here.
Food is similar where transportation doesn't matter much. Staple foods can be shipped by train, barge, and ship which reduces the cost. Some places are better at growing food. The canonical example is that New Zealand is so much better in growing sheep that more efficient to ship lamb by air. Growing local only makes sense for vegetables that need to be fresh or don't transport well.
TZubiri
Who complains about the land use of solar?
Obviously not a bottleneck? There's millions of desert acres.
ashoeafoot
My brothers farm had an offer by a similar startup with different dimensions but basically the same idea.
Had problems though. First was moving parts. You can not expect a system to last outside with any in them. Budget for the whole affair is tight, large repairs undo the whole endeavour .
Second is crops riping at different speeds below.
Finally practicality of the frame for farming machines . Meaning the dimensions need to adapt to pre-existing machines. And ideally remain upright even if one post gets hit.
Finally honesty regarding erosion ..corn is one of the more erosion prone crops and then having water dumped on specific areas concentrated creates channels fast. So the idea does not work with no crops or erosion crops in rainy areas. There is a reason there is grass below most solar fields.
One good aspect they didn't push is that this is ideal for electro farming. We have hyper effective electric moisture traps now and electro nitrogen fixing- combine that and this can remove one need for driving trucks through.
In all other aspects i see this limited to orchards.
otherme123
> In all other aspects i see this limited to orchards.
This study is for a very specific region: East Africa (Kenia and Tanzania). Their main problems are too much evaporation, no energy sources, no other sources of food.
I don't see how a solution that might work for a good share of humanity (Africa is huge: Kenya and Tanzania alone have 150 million people) gets dismissed as "limited to orchards" because it can't be used for industrial scale corn crops in my country.
rob74
Well, unfortunately "Africa" is only mentioned in the article's title (which got cut to fit into HN's character limit) and not repeated in the article itself.
But I can understand the concern: most agriculture (not only corn) is highly mechanized pretty much everywhere in the world except the poorest countries, so this would be limited to the crops that can't be mechanized.
otherme123
Or this can be adapted to mechanized crops where it makes sense. Greenhouses where small fragile minihouses of 4x3 meters, made of wood and glass 100 years ago, used only in countries with harsh winters. Nobody in their right mind would use a tractor to work on them. Today they can be seen from the space and are mechanized... where it makes sense to do it. They turned a desert into a huge production of fresh vegetables: https://www.amusingplanet.com/2013/08/the-greenhouses-of-alm... .
Agrivoltaics could be something or nothing. But don't dismiss it because it today isn't already perfect for everything. Ten years ago the cost of the panels alone would make this projects 100% infeasible.
usrusr
> Finally practicality of the frame for farming machines . Meaning the dimensions need to adapt to pre-existing machines. And ideally remain upright even if one post gets hit.
I believe that the end-game for agrivoltaics must be a reversal of that relationship: not the panel scaffolding adapting to farming machinery, but farming machinery and panel scaffolding becoming one, the scaffolding doubling as rails for overhead machinery. The status quo in farming is that a lot of fertile ground is wasted on machinery tracks. Machinery is either narrow-wheeled and ruining the soil in the track through compression, or the wheels are very wide (for weight distribution) to cause less harm per square inch, but harming proportionally more.
When you have scaffolding, scaffolding that is strong enough to survive a storm or two, it will also be strong enough to carry machinery. Not the machinery you'd attach to a 600 HP tractor, but the entire incentive situation for machinery size is based on the amount of harvest lost to machinery tracks and that would be completely solved through scaffolding-based machinery. And the issue of machinery tracks (and ground compaction) only gets worse when you start considering decarbonization: batteries are heavy, and a grid connect right above your field would be just what you'd not even dare dreaming of.
Lutger
Vertical bifacial agrivoltaics system are a promising setup that addresses some of your concerns. Most importantly, tractors can easily pass through them.
Another good setup is to combine it while grazing sheep.
lostlogin
A problem with sheep is that they aren’t worth anything. Here is New Zealand we have replaced nearly all of them with cows. Much improved profits but destruction of the waterways. We call that a win.
Lutger
Good point. I'm not really interested in farming animals, I really don't know enough about it. In the Netherlands we still slaughter little under a million of them per year, on a populations of circa 17 million that should still mean they are worth something.
But maybe grazing isn't worth anything at all and its only cost-effective if you put them in a big box and feed them grains or something. I don't know.
the_sleaze_
Hot take: compress them all cheek to cheek in a warehouse, feed them corn and capture the methane they produce. Cover the roof in solar.
Call that an even bigger win...
lurk2
> One good aspect they didn't push is that this is ideal for electro farming. We have hyper effective electric moisture traps now and electro nitrogen fixing- combine that and this can remove one need for driving trucks through.
Could you tell us more about this? I remember seeing a user comment a few years ago discussing the prospect of dumping excess electricity into nitrogen production but I had assumed it wasn't feasible. Does this technology really exist?
ashoeafoot
https://advanced.onlinelibrary.wiley.com/doi/full/10.1002/ad...
There is one experiment in kenia
ashoeafoot
https://www.youtube.com/watch?v=G2brxBRnRH4
high voltage moisture trap
cmrdporcupine
Definitely doesn't seem appropriate for cash crops harvested by combine harvester.
Most orchards and vineyards in temperate climates are attempting to maximize solar exposure, so would compete for space with the panels -- and also have their own trellising systems and so I struggle to see a fit there, too.
But I can definitely see the use cases for two places.
Grazing land for some ruminants, especially in hot dry climates. Provide shelter for animals, and shade for grasses.
Vegetables, market garden or even large scale. (e.g. my neighbour grows something like 20 acres of cauliflower and it's all planted and picked by hand.) Many market gardens are using walk-behind / two-wheel tractors (BCS, etc.) which are far more agile and could easily handle moving around panels.
Spill-off / erosion can be dealt with through building swales.
ashoeafoot
actually , the orchards need constant sun is a myth too. There is a very sharp cut going towards too much sun, starting near sicily and travelling north.
https://citrusindustry.net/2022/07/05/protecting-citrus-tree...
eldaisfish
that's ok. Not every solution must work for all situations.
the point here is that agrivoltaics tends to work well, especially in certain conditions. In central Africa, those conditions are reducing evaporation and providing shade to certain plants.
In north america, those conditions are often sheep grazing, as is common in New York and Ontario.
cmrdporcupine
Aside: There seems to be an uptick in sheep rearing in my area (southern Ontario, near Hamilton) all the sudden. In the last year, I've seen 3 farms local to me put up fencing to start grazing them. Which is interesting because it's actually very difficult to find Ontario lamb in the butcher shop, and it's not a very popular meat overall, so I'm wondering where the source of the demand is. The majority of the lamb in the local grocery stores is either frozen prepacked from New Zealand, or from Sun Gold foods in Alberta. But I find local lamb from this area tastes much better (less gamey), must be something about the grass vs whatever forage they're giving them in Alberta.
My two border collies would love it if I raised sheep. But my quasi-vegetarian wife could never tolerate raising them for meat.. and dairy and wool make no $$ sense.
tomrod
Or hayfields.
grumpy-de-sre
I really don't get the hype around Agrivoltaics at all. It all kind of pivots of the idea that land for Photovoltaics is scarce. Maybe it is in the UK but in the vast majority of the world this is a non sequitur.
Cover the crop in shade cloth, and mount the panels on the ground / roof structures nearby. No need to over complicate things.
closewith
> Maybe it is in the UK but in the vast majority of the world this is a non sequitur.
Land near energy demand is scarce almost everywhere. This kind of local generation is ideal and both reduces transmission losses and makes the grid/society more resilient.
mapt
Land near energy demand is not scarce almost anywhere. Where "near" means "Within 1000km". And it doesn't need to be near. Transmission lines are cheap and highly efficient infrastructure if you have a sensible process of land tenure (whether via routinized/generous eminent domain, Harberger Taxation, or other means).
Moving electricity is among the cheapest, most efficient industrial processes that exist.
*With all due respect to the Maldives, Gaza, Hong Kong, et cetera
wongarsu
At least in Europe, building transmission lines is a much longer process than building solar. Partially because of the complexity of acquiring land from a large number of owners, but more importantly because of NIMBYs that do everything to oppose new transmission lines that might disrupt the landscape. And some of the NIMBYs own the land you need to build the transmission line. Building a 1000km high capacity transmission line is a multi-decade project in this part of the world
grumpy-de-sre
And don't be sleeping on the potential of UHVDC transmission.
closewith
To realise the benefits of local generation, near means on the order of 1,000m, not km.
blitzar
Roofs near energy demand is plentiful almost everywhere.
I am lazy, I like to start with the low hanging fruit first.
closewith
> I am lazy, I like to start with the low hanging fruit first.
Then you should be looking at agrivoltaics before rooftop, as they are much cheaper in developed countries where the cost of rooftop solar is dominated by installation cost.
grumpy-de-sre
Rooftop photovoltaic can (and will) cover local needs. Bigger issue is local energy storage (eg. community batteries).
closewith
This is absurd. There's no limit to energy needs - they only grow.
aa-jv
Do you have your own personal garden?
I'm excited about it because its a combination of many things I truly enjoy doing and get a lot of pleasure from - harvesting energy, growing food, seeing life in equilibrium.
The integrated growth-stands with full-spectrum solar -> plant -> fish -> soil process is really, really appealing. Healthier food, energy harvested in neat ways. If I can get carrots and radishes and salads from the same device that charges my cell phone in the next year or so, I'll be a happier human for sure.
grumpy-de-sre
> Do you have your own personal garden?
Yes, every season I grow ~20kg of vegetables on a 15m^2 balcony (tomatoes, cucumbers, peppers, celery, various herbs). They are grown vertically on a trellis system, using 11 liter hydroponic "bato" buckets and recycled/reused perlite (sterilized). Nothing fancy.
The main energy input is the chemical fertilizer. Electricity to run pumps/valves etc is very minimal and probably amounts to 5-6 KWh over the season. I use synthetic fertilizers but avoid insecticides/fungicides, instead opting to use biocontrols (shout out to Koppert! https://www.youtube.com/watch?v=AFPkAQUfYvo).
The future of agriculture (excluding cereal crops, which I dream might be replaced by synthetic starches one day) probably looks something similar to: https://www.youtube.com/watch?v=a_pR_HihCVo
barrenko
Now this is solarpunk.
andy_ppp
Yes you could easily combine solar panels and poly-tunnels for example. We have the potential over the next few years to make everything work so well but instead the US seems to be cutting cyber funding.
ZeroGravitas
I've read an article with vertical bifacial panels between long hemispherical greenhouses to catch reflections.
grumpy-de-sre
Photovoltaic "fences" are one of those bizarre ideas that might actually make sense.
rompic
Just read an article about one of these in the local newspaper.
They currently have some challenges with bureaucracy, at least in Austria:
Given that it is an industrial facility it has to be secured from unauthorized access, but as it is a field, it has to be accessable to small animals. So now it is fenced with barbed wire on the top but it's open at the bottom.
tfourb
Not the case in Germany. Photovoltaic installations on agricultural plots are routine, in certain cases (I.e. the plot is next to a motorway) you don’t even need a planning/building permit, as long as the installation conforms to established building practices. Agrivoltaics is still a niche practice, but not because of bureaucratic concerns.
metalman
so called agrivoltaics will have certain very limited applications, and putting pv panels on the many very large agricultural buildings roofs, is much easier to integrate. A big risk with "agrivoltaics" is that they will in fact destroy the agricultural potential of land and then become the only source of income for the "farmers" , therby converting protected agricultural land, into purely industrial uses. Putting solar pv over canals is a better option, it reduces evaporation, while providing a cooler operating enviroment for the panels, in places that have no real constraints with developing them further.Also resevoirs can benifit/help in the same way with floating solar pv.
KaiserPro
> putting pv panels on the many very large agricultural buildings roofs, is much easier to integrate
for industrial farming yes. This study is aimed at subsistence farmers in africa. ones that don't have much mechanisation.
> will in fact destroy the agricultural potential of land
Given that this study highlights the water conservation features of agrivoltaics, I'd say it has a greater chance of stopping salinisation of agricultural land from over irrigation.
> Also resevoirs can benifit/help in the same way with floating solar pv.
again, this is aimed at african farmers with limited infrastructure. this would be the stepping stone to getting cheap electricity
thinkindie
something similar came out a couple of years ago for growing lemons in the south of italy with solar panels providing shades from an ever increasing temperatures https://www.bbc.com/future/article/20230424-how-agrivoltaics...
phr4ts
If this is target at developing nations, theft of the solar panels will be a major hindrance.
leemelone
This is a pretty dumb idea. This article is from researchers at a University. Has anyone done this for real without a grant subsidy?
I understand the interest in maximizing space utilization, but this just doesn’t make any sense.
Some replies indicate field setups for solar should take priority and tractor/equipment systems should adapt to these new systems.
Crazy stuff.
postepowanieadm
There's a simpler way: you may burn oats and generate heat/power without relying on problematic solar panels. It's already quite common.
rcxdude
Growing crops for fuel is so absolutely ridiculously inefficient it's amazing anyone does it at all (It's usually not even energy-positive: the only way it makes sense is if you want to make sure you have excess food production and have something you can do with the waste).
eliaspro
Comparing the landuse efficiency of corn-based biomass vs PV is about 60-75x.
It just doesn't make any sense at all to use crops for energy production.
pjc50
The only case where it might be viable is burning crop waste; stems of grain and maize, for example. The UK runs Drax on (imported! in oil burning ships!) sawmill waste, but that's greenwashing to me.
eliaspro
It definitely is. Besides that, Drax doesn't only run on waste products (that's what they'd like to make us believe), they mostly run on freshly logged timber.
Most of it, imported from the US: https://www.biofuelwatch.org.uk/wp-content/uploads/Drax-brie...
And even if it only was waste that's burned, it's IMHO incredibly stupid. We should be happy about every single gram of CO2 that's stored somewhere and use this "waste" for building insulation, paper production, soil revitalization or simply bury it.
Burning simply shouldn't happen when we have way better, cheaper, more efficient, more environmentally friendly methods of producing heat and electricity.
pfdietz
In a post-fossil fuel age, that crop waste will be used as a carbonaceous feedstock for things like jet fuel and what are now petrochemicals. Burning it for power will be uneconomical.
usrusr
Unless you start considering carbon capture: if some direct carbon capture folly exists and you replace that folly with capturing the output of some crops-to-energy setup, then that would surely be a massive improvement over the carbon capture it replaced.
pinkmuffinere
Sorry, how is this better? The whole path of [grow oats -> maintain steam power plant -> ship and burn oats] seems much harder to me than using solar cells?
ZeroGravitas
What's problematic about solar panels?
There's a reasonable amount of waste biomass generated by farms that can be used in various ways without growing it specifically for that use.
postepowanieadm
Mainly recycling.
dukeyukey
And the problem with that? Seems we can reasonably easily recycle the vast majority of solar panels, no differently to any other electronic good.
ZeroGravitas
Even if you couldn't recycle them, they'd be a much better solution than what they replace.
The fact that you can recycle them just makes them even better.
pjc50
People just like using "problematic" as a content-free slur that's acceptable in "progressive" circles that bypasses critical thinking.
KaiserPro
Oats don't like hot dry areas, such as kenya in summer/dry season.
plus, burning oats is really fucking expensive, the chaff is a reasonable feed, and the straw is good bedding/binder for building stuff.
Burning for power is really a poor use, especially as it needs compressing into bricks first. Plus you only get a tiny bit of energy once a year, compared to ~1kwhr per panel per day. (depends on panel size and positioning etc etc )
looofooo0
No, EROEI for this is just bad if not negative. Also you do not produce any food.
thowawatp302
What does “problematic” even mean here?
Agriphotovoltaics is largely a solution for a non-existing problem.
It is always brought forward as an argument when the perceived landuse issue of solar is brought forward. But if you do any serious analysis, the landuse impact of even utility scale solar is so neglegible that it's not an issue to worry about.
If you want to talk about landuse impacts of renewable energy, there's really only one issue to focus on: bioenergy crops. It can be as bad as 30 times less space efficient than generating the same energy with solar photovoltaics.
Talk about using the same space to make food and bioenergy. Use agricultural land for food, and utilize the residuals as good as possible. Stop making biofuels out of food. And treat the space needed for utility-scale solar as what it is: tiny.
(P.S.: If people can make agriphotovoltaics work, I have no issue with that. Great. Even tiny impacts should be reduced if possible. But I have an issue with the way this is discussed, because it makes up a problem with photovoltaics that largely doesn't exist.)