Wheat yield potential in controlled-environment vertical farms
pnas.org
pnas.org
"With artificial lighting increasing the intensity and duration of light beyond what can be captured from the sun in a field, the short indoor growth cycle produced mean grain yields of 14 ± 0.8 t/ha per harvest at 11% grain moisture based on a 1-m2 edge-protected experimental area"
How tf can you say your yields are going to extrapolate from that tiny space to a hectare scale facility? It's ridiculous. We have a huge problem achieving lab-theoretical yields on working farms, outside of the super optimized and most destructive conventional agriculture methods. They didn't even do a full greenhouse trial. Come on guys, you can say it's promising but to say you can get 1000+t/ha out of a vertical farm because of this is fantasy.
Ostensibly you would be able to do all sorts of shenanigans with light tubes for natural light and LEDs for artificial lighting regardless of the size of the facility.
The whole point of this paper is about having high intensity artificial lighting on the crop. So-- I'm not sure what your comment means.
I can understand the argument "it isn't economic", but the process to make something economic is do it once -> do it many times -> do it many times cheaply, so the paper could be part of the process that ends with mass vertical farming at absurd yields.
And that would be true, but it could still be a good target within an order of magnitude, because it could be set up as a very big RAID array. There would losses due to cooling and power and space between the drives and whatnot, but to a first approximation (Facility Volume / 1Tb density * 50% fudge factor) will be pretty reasonable; especially if someone really tried to engineer clever heating solutions.
So the paper gets (checks) ~1,300t/ha lab conditions. Why, in theory, can't they just replicate the lab environment as many times as fits in a big facility? I can see practical impediments like cost, I can't see theoretical ones if it is decided that This Must Be Done. I'd expect reasonable yields of 650 t/ha. That improves on 17 t/ha farming practice [0]. Point is that sort of calculation is totally routine when looking at academic results and I don't see what the problem is using a 3'x3' lab environment then extrapolating with a fudge factor.
As a complete outsider, I say scale it out rather than complain! I eat chapati every day, so I am biased towards wanting success to occur.
It is always possible that endeavors will fail.
"Those who say something is impossible should get out of the way of those who are doing it"
Light, nutrient and water distribution can be replicated, but making sure things scale properly and you don't get cascading failures is key.
Vertical farming is a pipe dream.
"However, given the high energy costs for artificial lighting and capital costs, it is unlikely to be economically competitive with current market prices."
There is a limited amount of sunlight hitting an area. That means to scale crop yields by going vertical, you need artificial light.
At the moment energy is not harvested in a (sustainably) scalable manner, so vertical farming is infeasible except to boost local production in rich countries.
Saying it's a "pipe dream" is a bit too harsh imo, I think as energy demand ramps up, and oil keeps going down eventually we will be forced to produce unlimited amounts of cheap energy through nuclear power plants, but that's just my vision for the future..
2. I’ve heard the real efficiency is tailoring light from high efficiency leds per plant.
If we do that it would nice to have some experience ready to go on the Industrial farming side.
Also it takes protons to haul the wheat to market, to harvest it, til the soil and plant it. Not to mention all the externalized costs of large scale outdoor agriculture.
Also on mars and the moon beckon, having a closed loop there would likely dwarf anything we can grow on a field here because of the energy costs of moving tons of food into space.
This is like saying that you can grow X goldfishes in your tiny little aquarium and therefore it should be possible to grow X * FACTOR in a lake which is FACTOR size bigger.
The same applies to open farming vs growing something indoors.
Now coming to this example about vertical farming - at much larger scale you will not get the sort of yield/sq.ft you get at a much smaller scale - even in indoor farming.
Reason one, is that I think one lesson we've learned and are still learning is that the original model of the internet that senses damage and routes around it is still a strong model, and I think the same applies to food. To reduce the impact of food shortages, we need to get more people farming again essentially, and a small vertical farm can be had for those people regardless of if they have land or suitable land for traditional growing, and the cost are likely to be similar to solar panels and be one that is subsidized at first or at least given tax breaks.
Reason two is that I think the same work will be of vital importance for future space faring missions. If we can get closed loop ecosystem sustainability heavily improved we might be ready for something like a real Mars trip/settlement etc. Personally I advocate for adding other systems to the closed loop to feed off each other, aka auqoponics, etc to add a meat source (fish) into the loop etc.
So we have a decent study on wheat. Now lets get the yield potentials for everything else under the sun so we can start optimizing better. Probably end up with focus on whatever has the highest calorie output to price to produce ratio.
https://www.youtube.com/watch?v=wsaufB5F8dk
That presentation is by one of the same scientists referenced in the linked paper, Dr. Bruce Bugbee, the one who grew wheat on the ISS.
He has a whole youtube channel that goes into more detail than you can imagine: https://www.youtube.com/c/Apogeeinstrumentsincorporated/vide...
I wonder if "packetizing" food production in a vertical permaculture food forest form factor is feasible? In other words, instead of vast monoculture swaths, use interdependent crops in interlocking planting patterns, predicated upon the assumption that robot-vision-assisted labor is practical.
If it's a practical business model, it should be practical somewhere in the world simply using existing human workers.
Basically, you put them on a hill :) Now you have an underground farm at high ASL. Problem fixed.
For every $10 of wheat produced it costs -
$X for gas for production $X for transportation and packaging $X for human labor $X for machinery $X to buy the land etc.
I'm guessing that land ownership is a tiny minority of this - and if you are building towers that are 50x more expensive than an empty field it would make the process more expensive and less energy efficient if you factor in upfront costs. Making land more efficient is great but I don't think it is the limiting factor for supplying ample food for everyone in the country or the world.
I think you’re missing my favorite point: we can return land to the environment and potentially reduce the entire effects of pesticides.
Even if it were less efficient, that would be worth something.
I am not able to see how a network of distributed vertical food farms would be less resilient to political disasters than growing everything in the Midwest. Food production has been the foundation of every successful civilization, this technology would potentially allow us to grow food without wrecking Earths environment.
California and several other states are also net food exporters.
Farming enough grapes (as my late Uncle did in California), say, to balance trade doesn't mean you could farm enough food in general.
Both USA and UK recently stopped funding pandemic preparations.
Keeping spare capacity on hand when the money could go to private profits instead .. not with our current political systems.
If we can forcibly spread the benefits of technological progress, and require long-term planning in the political processes then we (humanity) can probably do this sort of thing.
I fixed it for you. States like Indiana import 90% of their calories. Midwest farming is all about massive row crops, not human food.
https://www.crcworks.org/infood.pdf
Yet Indiana does not even feed itself, let alone feed the world. The state imports an estimated 90% of its food. More than $14.5 billion is spent by Hoosier consumers each year buying food sourced outside of the state.
When you consider that many countries import energy through natural gas from Russia vertical farms could be shut down during a war all at once and force the affected country to surrender immediately.
Mind you Western Capitalism loves consumer wastage as it leads to higher sales, wastage is designed in.
Vertical farms and current political systems just means more wastage will be encouraged and more obesity generated with less expense at the front end.
Citation? Sounds good in theory but many problems could arise, such as molds, fungus, or complete wiping out of a monoculture from disease.
"The expansion of agriculture has been one of humanity’s largest impacts on the environment. It has transformed habitats and is one of the greatest pressures for biodiversity: of the 28,000 species evaluated to be threatened with extinction on the IUCN Red List, agriculture is listed as a threat for 24,000 of them"
https://ourworldindata.org/land-use
Do we really want to keep eating into what wilderness remains?
Just because we "have" it doesn't mean we should convert it.
We already produce enough food to feed the world and likely won't have trouble producing enough food to feed the coming extra billions, but if we can focus on increasing yeilds and moving away from raising animals for consumption we can also reduce our footprint on the world.
If we want to release cropland back to the wild, then by far the cheapest way to do it is improve farming infrastructure and practice in low income countries, which is where yields lag so far behind the state of the art.
Not a technical problem, a social one.
And who in those countries gets the advances? A corporation or a community - if a community currently grow the own food inefficiently and you take the growing land, give it to a corporation and produce the same food efficiently then the people lose income, need to buy their food, and all financial benefits of the food sale get privatised. The value produced by that land is moved. We need to be careful not to spread the bad parts of efficient production.
I saw some mango farming videos in Africa and that's pretty much it. There are lots of inefficient farming practices and agriculture experts are teaching farmers how to improve their yields significantly without any fancy technology.
Much of US produce comes a few counties in California 100% dependent on the Colorado river — what are the risks with respect to climate change on reduced snowfall, etc in the Sierras? Midwest irrigation depends on aquifers that are being irreparably depleted.
On the east coast, the dominance of California produce and decline of dairy killed many forms of agriculture. The endless suburbia of New Jersey was once prime farmland, gone forever.
I’m pretty sure that will always be the ratio, and there’s nothing anyone can do to improve it.
Even if we could only implement vertical farming at a similar level of productivity to traditional agriculture on a horizontal basis (although it seems from the paper that we could do a lot better than this), with a unit height on the order 1 metre/unit, we could still achieve productive densities hundreds if not thousands of times higher, given sufficient power and building technology.
I can see perhaps one order of magnitude from scaled-up manufacturing, but where do the others come from?
Wait...
https://www.scientificamerican.com/article/plants-versus-pho...
Chewing on bark will work, if you have enough bark, but it’s significantly more effective to use a fungus as an intermediary.
From the appendix¹ (assuming I am reading this right):
Value of production 405,329$/year
Cost of energy 15,987,286$/year
Area of field for equivalent yield 6,333,272m²
Land Area Required for PV Array 4,157,293m²
Of course, this obviously depends on you making money on the wheat or other crops.
First, plants lose about half the energy because of spectrum. There's lots of other losses (e.g. light falling on parts of the plant other than chloroplasts) but they apply to artificial light, too. And then, plants use only about 10% of the mid-day energy after all of these losses, because of saturation effects, but more the rest of the day-- we'll call it like 25% here. So that 50% factor and 25% factors multiplied are what you can try to exceed == 12.5%.
Cheap solar panels are about 15% efficient. So you're not going to beat direct sunlight by much.
That comment is using numbers that are overly optimistic. Using 22% panels requires you to construct an unaffordable PV facility for $400 million for a single vertical farm. If you go with lower efficiency panels you might be able to make it cost effective but then you are also consuming more land than a conventional farm. There is also a mismatch in energy demand. Solar panels only work while the sun is up but the yield promises of vertical farms entirely depend on 24/7 lighting.
The cost of steam turbines and generators alone would prevent nuclear plants from producing power at 1/10th the cost of current plants. I don't think you're going to find a nuclear component of the plant that costs negative dollars.
[1] https://whatisnuclear.com/economics.html#early-days
[2] https://m.youtube.com/watch?v=Rb7tAwNIvUs&feature=youtu.be
These turnkey contracts cost the nuclear vendors a lot of money, and they moved away from them for that reason. Only in the last wave did Westinghouse go back to these kinds of contracts, and it drove them to bankruptcy.
regulations were changed retroactively as well, for already approved projects.
Ill look around for better links to clarify this.
And I explained why that was misleading. They were being sold at a loss. The PRICE was competitive with coal, but the cost was not. This was even before changes in regulations.
Can you point me on where to look for the price/cost distinction in early nuclear? I was not able to find definitive info there.
https://www.theatlantic.com/technology/archive/2011/03/the-n...
See also
The Turnkey Era in Nuclear Power, H. Stuart Burness, W. David Montgomery and James P. Quirk, Land Economics Vol. 56, No. 2 (May, 1980) (available through JSTOR)
The latter is interesting, as it also gives an explanation for why utilities might have liked non-turnkey contracts: any escalation of capital cost can be (and very often was) passed on to ratepayers. In light of this, what really nailed the first US nuclear age was the passage in 1978 of PURPA, particularly the part about non-utility suppliers.
https://en.wikipedia.org/wiki/Public_Utility_Regulatory_Poli...
As for turnkey contract, interestingly, other industries trying to avoid uncertainty and not pass costs to consumers.
AFAIK, Russian nuclear contracts come with warranty of all funding, so if overrun happens, these are Rosatom problem, not customer's
My point is that with cheaper than coal it becomes viable co2-less source of energy for,among other things, vertical farming.
it will allow growing regardless of the weather and clouds above.
Building a close to the city, colocated flour, and/or cereal factory under the wheat farm, allowing gravity to help build automations would reduce holistic costs. I wonder if that makes this farming method profitable?
Immediate second order effect would be the obsolescence of grain futures, you don't need to hedge against a bad harvest. You probably also wouldn't need to go into debt to buy enough seed to plant each season.
Another would be a more elastic supply chain without the need for storage/buffering, which we've seen drastically reduce prices and increase production for all sorts of widgets using just-in-time manufacturing. If you can scale up/down wheat production pretty reliably with only 70-80 days forecasting, you can eliminate a lot of the inefficiencies of the food supply chain.
Sidenote, a literal vertically integrated bakery would be a sight to behold
Growing a metric ton of wheat with 100% supplemental light requires 150,000-400,000 kwh. That's $3,000 to $8,000 in electricity costs at 2 cents per kwh (very cheap) and would require electricity to generate about a gram of CO2 per kwh at the most to be better in terms of CO2 emissions.
If you have free, non-polluting electricity it would be better to use it to synthesize fuel for trucks to carry grain from the fields, or to capture CO2 from the air.
In theory the energy intensity of bitcoin should go down over time because the block reward is shrinking. Of course transaction fees could rise but processing transactions is different to wasting energy to mint a currency out of thin air. Paying transaction fees is like paying your taxi driver. It's expensive but it's absolutely necessary to get a ride.
There are other inputs to growing outdoors: fertilizer usage, water usage, and insecticide/pesticide usage (likely not an exhaustive list). In a controlled environment like an indoor farm the use of each of those is dramatically less than in traditional farming. It would be worthwhile to include all of the inputs of growing food since light and transportation are not the full list.
(disclosure: I work for an indoor, vertical farming company)
Some back of the envelope calculations: From [1] the cost of fertilizer is like $150/acre and from [2] you can get about 7 tons/acre, so it's like $40/ton. The numbers change a lot from source to source, so let's multiply that by 2, and we get $40 of fertilizer per ton.
So in the impossible best scenario where the indoor production saves you the 100% of the fertilizer, you save $40 per ton of fertilizer that is much less that the $3000 per ton of electricity for illumination.
[1] https://farmdocdaily.illinois.edu/2017/07/fertilizer-costs-i...
[2] https://www.seedcorn.com/resources/estimating-corn-silage-yi...
As you said, simply finding ways to reduce the cost (in $ and CO2) of transporting field grown wheat is probably better than co-locating a vertical farm.
[0] https://www.statista.com/statistics/675822/average-prices-us...
The sun provides about 300W/m^2 on average at temperate climates, that's 7.2kwh per day. Wheat takes about 4 months to grow and nuclear power costs 0.77c/kwh.
That's 665.28$/m^2 at wholesale.
[0] https://ag.tennessee.edu/solar/Pages/What%20Is%20Solar%20Ene....
[1] https://en.wikipedia.org/wiki/Electricity_pricing#Price_comp...
But this would probably only halve theoretical costs, and then there are all the losses in the power production chain...
To grow 1000 pounds would require an area 26 ft by 26 ft.
For a product worth less than $1000.
Counting space, heat, light, equipment - doesn’t sound like a paying proposition.
Maize has even been intentionally hyrbidized for almost a century; I'm less familiar with wheat from that sense.
Yes, tools like CRISPR are game changers and should be approached with caution (and we almost certainly aren't going to do so, at least for animal feed-stock crops), but GMO labeled food is very broad brush with often arbitrary distinctions (much like the "artificial" vs "natural" labeling rules)
But look at a picture of teosinte, and compare it to maize. Maize is emphatically not a natural plant; why should a plant grow appendages it's not strong enough to support? That shift in plant structure occurred over perhaps one or two millennia, and would have required extremely active selective breeding to occur.
Start poking at our plants' genomes, and the evidence for calling it genetic engineering is somewhat clearer. Our cereal crops basically suffer from plant equivalents to genetic conditions such as Down Syndrome. They shouldn't be able to survive, and indeed generally cannot survive without active management by humans. And being unable to survive without human intervention is common enough that we've given it a special name: domestication.
Scenario 1):
A random mutation decreases my yields. My community of 30 people can barely get by on the food we grow. Some of our children died from starvation and we are down to 28.
Scenario 2):
A random mutation increases my yields. My community of 30 people can easily get by on the food we grow. None of our children are dying and we expect 5 children to be born this year. I can sell the extra seeds at the farmer's market and make an additional profit.
Absolutely zero genetic engineering is needed to improve yields. This is just natural selection. Nobody has to be aware of the process for improvements to happen.
The next step would be to just buy the best seeds available at the farmers market. It would significantly more effective than natural selection but all you're doing is buying seeds from the farmer who is selling more food than everyone else.
Calling this engineering would be insulting to every other branch of engineering. Anyone buying food would be genetically engineering their food by this definition. Seriously, what other branch of engineering requires no knowledge of it happening?
By that same logic everyone is an economist. The laws of supply and demand happen even if you are not aware of them.
I'm tired of people diluting "professional" words into meaninglessness. It's not a colloquial term and it certainly shouldn't become one.
People would barter and buy seeds from farmers with higher yields with the intent of getting those larger yields themselves in future years. This is not natural selection.
I'm equally sick of people implying that ancient people were stupid and incapable of engineering.
As you mention fields, they are in fact an ancient invention to allow farming, and most importantly, an artificial invention. Nobody gets a full belly from the 3 plants in a window pot. One has to do large scale agriculture to feed large numbers of people, and this has been conducted for millenia. A field is at the very basis of agriculture. In order to make land arable, we've cut down forests, moved stones, drained swamps, leveled terrain, built irrigation systems, etc. Basically engineering work.
Fields do look like nature, but they don't represent nature. Wild type nature looks different.
TLDR: Humans are engineers. Always have been.
Not continent-spanning grow-labs, which is the proposition being forwarded.
2) Seed will be contaminated. It will taste as leaf or straw.
See p. 12 of the Supporting Information and you'll find the following: 2026647 kg wheat grown using 798417 MWh electricity means an electricity consumption of 0.4 MWh/kg wheat. In the US, the annual wheat harvest is 150-200 kg/capita/yr, which in this hypothetical system would use more than 60 MWh electricity/capita/yr.
For comparison, the US annually uses around 13 MWh electricity/capita.
That is, this hypothetical wheat production in the US would use about five times the _total_ present day electricity use.
Soybeans and maize together occupy more than three times the wheat area in the US, so that (very roughly speaking) adds another annual electricity consumption of ~200 MWh/capita. With just these three major crops, the US would have to increase its total electricity consumption about 20-fold.
Just say no to vertical farming for row crops. Can’t beat the sun (and scale)!
one argument against vertical farming seems to be the high occupancy costs in larger cities. Why not build them outside of cities and connect them?
Could you explain what you mean a bit further?
Right now we have excess food production.
> Right now we have excess food production.
True. If this vertical approach produces wheat for less money, then it will take over. If not, then it won't... today. There's this whole climate change thing. People suspect that it's going to harm food production. Can we control this smaller environment well enough for it to still be productive? Better than we can control outdoor farms.
Your system of mirrors will occupy the same amount of space (or close to) as your farm would need. But it will be expensive to build and expensive to maintain and to protect against the elements.
The single most sane way to change agriculture for the better is to eat much less beef. hands down. simple as that.
Like nuclear?
I think addressing the issue of not having to succumb to the model that ensures 1/3 of all food produced is lost would do FAR more than that, recapturing the losses and creating alternative processes to create more efficient supply chains would ensure we could curtail so many more problems.
I say this as a person who has actually farmed, worked professionally in culinary (including vegetarian and vegan cuisine) and has a background in Logistics/Supply Chain for Auto-Multinational Corps.
So, if you think Conventional (chemical) Ag Grain/Vegetable monocrops (which is what most people eat) are really that much better for the Environment/Humans than a sustainable (ideally Biodynamic) Livestock raising model is better, you really have no idea what you're talking about--I've worked on both, too. Although I agree the US' consumes way too much meat in general, but there are much bigger problems to solve and changing a culture's palette is typically generational barring a massive catastrophe (think: War). I think many Millennial and Gen Z diets are moving toward plant-based more than their Boomer counter-parts, which I have some reservations about: the image of a 'vegan' toddler at the Farmer's Market I worked at during my apprenticeship comes to mind and I thought he was undergoing chemo-therapy he looked so unhealthy.
Also, I like the fact that these things are being explored, obviously for terrestrial applications the use-cases are limited, but experimentation for long-term Mars colonization will require in-situ crop cultivation if it is to be sustainable, and what applies to Vertical farming could help design the new container garden model. Vertical farms are essentially playthings for hobbyists on Earth, even a Community garden with a moderate size greenhouse for off-season grows will yield (in every sense of the word) way more per/sqft with the exception of perhaps the now defunct business models: selling micro-greens to fine-dining restaurants in a post COVID World.
Detroit showed how the solution to Food deserts isn't Container Gardens/Vertical farming, though its not entirely against it either, but its a reversion to Agrarian practices which include re-purposing large plots of land to Urban Farming, which created a Renaissance of sorts in the last decade. I guerilla gardened as an activist before farming, and while vertical farming should be encouraged; if nothing else as a form of Community building and as small step in CO2 sequestration as well as an improvement aesthetically speaking.
Personally speaking: As a person who grew up in CA the Valley has to be the ugliest part of CA to me because of how modular and cookie-cutter everything looks, it looks like an 'Industrial Model' to Civilization.
A few creeping grape/pea vines and spontaneous gardens won't solve the massive Homeless issue that ones associates with the vista of the Valley but it could really help improve the overall feeling of the place: I seriously had to flee to Sonoma every chance I got to keep from going crazy when I was there as it looked and felt so alien to me.
Honestly, Labour and long-term Capital is the only thing stopping it in my opinion, everything is pretty much there to make the transition. The former being the biggest thing to be solved can now be an opportunity for all the Extinction Rebellion kids and Greta Thunberg followers who have opted out of Schooling to 'put up, or shut up' if they're serious about this movement, and the latter can easily be done by removing all the Ag subsidies overnight.
Its entirely do-able, I did it when it was far less viable after my apprenticeship and came back to the US and converted a few organic farms after having managed an existing Biodynamic one for a year.
I really hope most coders/tech people switch over to Ag Tech after this pandemic is over and they reflecyed on what exactly they're applying their labour toward as there is so much needed disruption and opportunity in what is the biggest Industry in the World (Food) because we just got a real wake up call about how precarious our living situation is on Earth when we saw all those shortages during COVID.
And the only 'genocide' to occur will be from continual reliance and outsourcing to the CCP as we've done for everything else, and extinction if we don't modernize our Ag, Food Supply and Supply Chains to something actually sustainable instead of using current practices that deplete our soil, water sources and arable farm land with the seriously reckless practices you're advocating for.
Prices are not sustainable with Conventional Ag, and they were never meant to be nor have they been truly adjusted for the Environmental damage they the incur as a result of State intervention, and are instead masked by a multitude of subsidies (which keep farmers into a near debt-slave based servitude situation) and why the consolidation of Ag occurred by Chemical/Pharma corps (where I come from as a Biologist) in the first place, and subsequently why things like diabetes, heart disease and obesity kill more people in wealthy developed nations than anything else--which you can probably argue is a form of genocide, as in these Industries are targeting the above poverty class of the World into using becoming clients of either of its branches and wreaking havoc regardless.
To be fair, I'm not for phasing out conventional or even GMO practices over-night as some regions would simply not be able to grow anything because of the damage to the soil/environment, but instead we should stop incentivizing it and remove its subsides and allow it to compete on its own merits after having shifted those subsidies (perhaps even a significant reduction) for the next 20-30 years in the form of long-term, low interest loans to small, organic/sustainable independent farmers, Community Gardens and seed money/Angel Investment to startups in Ag.
Just the budget of the Department of Ag for 2019 alone is staggering [1] just the $23 Billion in discretionary funds of the total $140 Billion could revolutionize how food is sourced, produced, and delivered into a near post-scarcity level (which on a caloric basis we are already at) in our Lifetimes! We could put a real dent in Carbon capture, reclamation and soil re-vitalization, which could in turn be another way to capture Green house gasses, and help the environment further.
I really think most people take for granted how broken and needlessly, complex, wasteful and ultimately poor for all involved our Modern Food Supply system is and how dire it is to have to repair it.
1: https://www.usda.gov/sites/default/files/documents/usda-fy19...
So when it comes to vertical farming crops like wheat, it's worth looking at when that price drops to "low enough". Really what we are talking about is the cost of the infrastructure needed to perpetually produce enough energy for the life time of the indoor farm.
And you are right that cheap, plentiful, clean energy will have lots of other interesting applications. IMHO, that's going to be the story of this century. Solving this will produce a new economic boom very much like oil/coal caused last century. We can solve a lot of issues with cheap energy. There are a lot of things bottle-necked on energy cost. Improvements on that front unlock new applications.
Terrestrially, you're never going to get more efficient use of an acre of sunlight (~4 MW at noon) than you will with photovoltaics (~75% loss ~= 1 MW at noon) and conversion to artificial light.
But it could totally work on some future Trantor with flying cars and fusion engines.
* edit: another poster has dug into the appendix and concluded that you might be able to do about as good w/r/t land use with photovoltaics, presumably due to other factors affecting plant growth.
green plants don't use the middle, thus they are green, of the spectrum - i.e. they use only 10-20% of the sunlight. So, very efficient photovoltaics (or other type of solar powerplant) -> electricity -> only red and violet LEDs may even beat the Mother Nature in efficiency a bit. Add to that various optimizations - for example with the LEDs you aren't limited by the Sun singular position in the sky and can shine light from all the needed angles, like no more leaves being in the shadow of other leaves, thus increasing the photosynthesis throughput of a given plant.
Eg I’ll pay a premium for pesticide free, fresh salad from a Urban vertical farm. Wheat less so
If we're pursuing maximum nutrient and calorie output for minimum fertilizer and energy input, why look at traditional plants at all?
The only thing worth growing in vertical farms is branded "Vertically Farmed" artisanal products for the rich.
Surely you can see wheat is not great for this.
You still then have to put the world's most expensive wheat through the artisanal bread making process. How will this be profitable?