Energy and Vertical Farms

Vertical farming is hot.

The idea that we should grow food under highly-controlled conditions in urban high-rises is getting plenty of media attention, along with pleas for public and private investment. A new vertical farming book by retired Columbia professor Dickson Despommier came out in October, and Sting bought the movie rights.

Vertical farm promoters claim these buildings will solve every problem from farmland shortages to population growth. They seem particularly fond of futuristic renderings of glassy skyscrapers alive with plants bathed in electric light. The grow lights are necessary because most crops would fail under the meager ration of sunlight that could filter into the depths of a skyscraper.

Vertical farm concept by United Future digital design and marketing agency.

Proponents of vertical farming commonly say it will save energy by eliminating the need for tractors, plows, and shipping. Some even think that vertical farms can incorporate enough solar panels, wind turbines, and methane generators to be net energy producers. For the most part, vertical farms are more vision than reality, but a few real-life examples are appearing, offering opportunities for critical evaluation of proponents’ claims. In this post I examine one such example from a farm energy perspective.

Vertical farming: A real-life example

An article in the Milwaukee Journal-Sentinal caught my attention last July because it describes an actual aquaponics operation in Racine, Wisconsin, growing fish (tilapia) and lettuce in a former industrial warehouse.

An old JI Case building once used to manufacture plows for farm fields is being transformed into a dirtless vertical farm where fish and lettuce are grown in a symbiotic system.

The farm, in a part of the city that once was an industrial hub, potentially could produce the same amount of food as 40 acres of land without the use of pesticides or fertilizer…

Fish are fed in a warehouse in Wisconsin. Waste water provides a nutrient solution for lettuce produced under grow lights.

The article offers some numbers to consider:

• 10,000 square feet of warehouse space are currently used for production;
• 7,000 heads of Bibb lettuce are harvested monthly;
• a pound of tilapia is harvested for every four heads of lettuce;
• the monthly electric bill is $2,800;
• the farmers plan to expand to fill the 200,000 square foot warehouse.

Two-thirds of Wisconsin's electricity comes from burning coal (source: Union of Concerned Scientists. Click image for original report).

Photosynthesis is fueled by grow lights, since sunlight never reaches the lettuce. The lights run at night, when electric rates are as low as 5 cents per kilowatt-hour. At this rate, the farm is using about 56,000 kilowatt-hours of electricity each month (calculation). Most of Wisconsin’s electricity comes from burning coal mined out of state, at a conversion rate of about 2 megawatt hours per ton of coal. Therefore, providing electricity to this model vertical farm for a year requires the equivalent of 336 tons of coal (calculation) (220 tons of actual coal, plus nuclear, natural gas, hydro, and other electricity sources). If the operation expands to 20 times its current size it will eat up 6,700 tons of coal equivalent annually.

Then there’s the energy that goes into producing the feed. A lifecycle analysis of intensive tilapia production by Nathan Pelletier and Peter Tyedmers, at Dalhousie University, found that production of feed requires about 9 gigajoules per metric tonne, or 15 gigajoules per tonne of fish. That comes to 143 gigajoules of energy (calculation) — equivalent to another 6 tons of coal (calculation assumes 27 GJ/t) — for enough feed to keep this little vertical farm going for a year. If the planned expansion is completed then producing feed for the facility will demand about 116 tons of coal equivalent annually. Feed production accounts for 91% of energy use in intensive lake-based aquaculture, but that same energy investment amounts to less than 2% of the energy used to keep the warehouse-based farm running (calculation).

A platter heaped with eight pounds of coal demonstrates the quantity needed to generate enough electricity to grow a head of lettuce and a quarter pound of tilapia in a model vertical farm.

The newspaper article doesn’t mention how much energy it takes to heat and cool the warehouse. Tilapia grow best at temperatures between 72 and 90ºF, and lettuce grows best between 60 and 65ºF.  Both the fish and the lettuce could grow at about 70ºF, but the lettuce will start to bolt at higher temperatures and tilapia growth will slow at lower temperatures. If the fish and the lettuce are in the same section of the warehouse then the temperature must be maintained within a very narrow range, which takes energy. Heating and cooling account for the bulk of energy use in temperate region greenhouses, but probably require less energy in a warehouse filled with large pools of water.

Lumps of coal will be used to fuel the furnace of a small tomato greenhouse in Kentucky. Heating and cooling account for most of the energy used by most greenhouses.

By my calculations, producing a head of lettuce and a quarter pound of fish in this model vertical farm uses the energy equivalent of more than eight pounds of coal (calculation).

Worse than greenhouses

Back in 2000, Swedish researchers Annika Karlsson-Kanyama and Mireille Faist calculated the amount of energy used to get lettuce to the plate in Sweden’s industrial food system. They found that greenhouse-grown lettuce took more than 100 times as much energy to produce as field-grown lettuce (original report; see Tables 7 & 8). Storing and transporting lettuce took about the same amount of energy regardless of production method. For the field grown lettuce, storage and transportation accounted for two-thirds of the total energy needed to get lettuce to the plate. For the greenhouse-grown lettuce, storage and transportation accounted for just 2% of that energy investment.

Energy expenditure and return for a head of Bibb lettuce and a quarter pound of tilapia from three production systems. Systems compared are low input field and lake production, high input greenhouse and pond production, and an integrated vertical farm system using electric light. Expenditure (negative) is associated with the production (farm) and processing/distribution (post-farm). phases. Return (positive) is the metabolic energy derived from consuming the food. Click chart for enlarged interactive version.

Delivered to the plate, the field and greenhouse-grown heads took about 125 and 6,000 Calories, respectively. Eating the lettuce itself provides just 21 Calories of metabolic energy. If we are concerned about energy use in our food system, concluded the Swedish researchers, we should stop growing lettuce in greenhouses.

A food factory like the profiled vertical farm makes greenhouses look good. At more than 24,000 Calories for a head of lettuce and a quarter pound of tilapia (calculation), it consumes 3.5 times more energy than a greenhouse and pond, or 33 times more than a field and lake.

Replacing coal with switchgrass

Far from saving farmland, a food system that incorporates vertical farms fueled by farm-grown renewables could dramatically  increase demand for farmland.

An acre of switchgrass typically yields about six dry tons, shown here as a trailer load of round bales. This load could power a typical American home for 5 months but would be used up by 10,000 square feet of a model vertical farm operation in under 5 days.

Switchgrass, for example, has been proposed as a coal replacement for electricity generation. A ton of dry switchgrass contains about two-thirds as much energy as a ton of coal. Switchgrass typically yields about 6 tons per acre, so 84 acres could produce the 500 tons of switchgrass needed to generate enough electricity to keep the profiled farm going for a year at its current scale (calculation), and 1,770 acres would be needed to keep it going after the planned expansion. Add the 84 acres of switchgrass to the 10,000 square feet of warehouse space currently used, and the annual lettuce yield plummets from 366,000 (calculation) to 1,000 heads per acre (calculation). A typical California lettuce farm collects 50,000 heads per acre over two annual harvests.

Replacing coal with wind

Two wind turbines generate electricity for the Spirit Lake School District in Iowa. The first began operation in 1993 and has a 250 kilowatt capacity; the second has operated since 2001 and has a 750 kilowatt capacity. Each turbine cost about a dollar per watt of generating capacity to install, and generates less than a quarter of its rated capacity. If conditions are similar in Racine, the smaller turbine would satisfy half the farm's electricity demand, and the larger turbine would satisfy twice the farm's electricity demand.

Wind is probably a better renewable energy source than switchgrass near the shore of Lake Michigan, where the farm is located. Indeed, vertical farm renderings often feature wind turbines.

Hundreds of wind turbines have been erected across Wisconsin, with many more planned. Each typically contributes about a quarter of its rated generating capacity to the grid, due to the intermittent nature of wind. For example, a $780,000 wind turbine erected by Iowa’s Spirit Lake School district in 2001 has a generating capacity of 750 kilowatts, but has actually generated an average of 173 kilowatts since construction. Assuming similar winds in Racine, an identical turbine would generate more than twice as much electricity as is currently used by the model vertical farm (calculation). Another eight of them would be needed after the planned expansion, more than doubling the expansion’s $6 million price tag (calculation).

Of course nine wind turbines this big wouldn’t fit on the warehouse site, and Wisconsin’s emerging rules on wind siting will likely require a 2,500 foot setback between turbines and property lines, prohibiting construction of even one turbine in an urban environment. Wisconsin has an active anti-wind lobby which is raising concerns about safety, noise, bat kills, and other perceived drawbacks associated with wind turbines in close proximity to people.

Replacing coal with methane from waste

Gordon Graff's Skyfarm concept for downtown Toronto would recycle organic waste into methane gas to generate electricity.

The energy demanded by artificial lighting of vertical farms concerns Gordon Graff, a Toronto-based architecture student who designed a vertical farm concept called the Skyfarm. He estimates that his proposed 59 storey farm could grow enough produce for 50,000 households, but lighting it would require 82 million kilowatt hours each year — as much electricity as is typically used by 8,000 North American households. Graff proposes that half of this energy could be recovered by burning methane generated from the farm’s own waste. Based on the kilogram per kilowatt hour conversion rate calculated by the University of Reading’s Charles Banks, the Skyfarm would have to produce 45,000 tons of wet waste annually to satisfy half its energy demand through waste recycling (calculation). The remainder could come from methane generated from other municipal wastes.

Building biogas digesters to make electricity from municipal waste is a good idea, but using that electricity for grow lights to replace sunlight probably isn’t. If Toronto were to build the biogas digester that Graff envisions for his Skyfarm, but leave the rest of the project on the drawing board, the city would have 90,000 tons less waste for its landfills every year, and 8,000 fewer homes to power with coal and nuclear energy. If the rest of the Skyfarm is built, the city would have to resume landfilling 45,000 tons of waste and return those 8,000 homes to coal and nuclear energy.

Replacing coal with solar

Vertical farm renderings often feature opaque roofs covered with solar collectors. Solar dish collectors can convert about 30% of the energy they receive from sunlight into electricity. High pressure sodium lamps — the most efficient grow lights available — convert 12-22% of that electricity back to light. Therefore, a good solar system used to power electric lights produces less than 7% as much light as it collects (calculation). This contrasts with a clear glass roof, which transmits 90% of incoming solar radiation to the plants below. Spread over 10 storeys, a rooftop solar collector used to power artificial lights would do well to provide 0.7% as much light as would be available to a plant growing on the roof. A rooftop solar collector is an expensive and ineffective substitute for a glass roof.

Not organic

Vertical farms are often described as organic because they do not use pesticides. Organic farmers are quick to point out that organic agriculture involves much more than simply avoiding synthetic pesticide use: The term refers to farming systems that build soil organic matter to feed the diverse array of beneficial soil organisms that interact with plant roots. According to the USDA, organic farms respond to “site-specific conditions by integrating cultural, biological, and mechanical practices that foster cycling of resources, promote ecological balance, and conserve biodiversity.” In contrast, vertical farms wall themselves off from nature and attempt to replace free ecosystem services with energy-intensive engineering.

Whether soil-less production systems can even qualify as organic is controversial. This April, members of the National Organic  Standards Board voted 12-1 to recommend that the following clarification be adopted by the USDA:

Observing the framework of organic farming based on its foundation of sound management of soil biology and ecology, it becomes clear that systems of crop production that eliminate soil from the system, such as hydroponics or aeroponics, cannot be considered as examples of acceptable organic farming practices.

This language will likely be adopted as part of a new standard for organic greenhouse production, outlawing use of the word “organic” in the USA to describe terrestrial plants grown hydroponically.

Organic aquaculture standards are also mired in controversy. The National Organic Standards Board has recommended that synthetic amino acids and protein from wild aquatic animals be prohibited in organic feed. If enacted, these rules could increase the energy used to produce organic feed. Adoption of organic aquaculture standards is favored by groups like The Environmental Defense Fund and the Ocean Conservancy, but opposed by the likes of Greenpeace and the Sierra Club.

Organic certification of a system like the profiled aquaponics operation would require approval of separate organic farm plans for the tilapia and lettuce components of the farm. At present, neither would be approved. There might be hope for the tilapia component, but it doesn’t look good for the lettuce.

Local, not sustainable

The energy required to grow plants under artificial light far exceeds the energy  required to transport food from fields fueled by sunlight. A recent analysis of the US food system by a team of USDA economists led by Patrick Canning found that transportation is the smallest segment of the food system’s substantial energy demand. In general, growing food uses more than three times as much energy as transporting it. Vertical farms dramatically expand the agriculture slice of the food energy pie in order to reduce the much smaller transportation slice. They do not reduce the amount of energy used to process, package, sell, store and prepare food, which accounts for more than 80% of food system energy use.

Energy used by components of the US food system in 2002 (right), relative to total US energy use (left). Data from Canning et al., 2010 (Click image for report).

Dreams dashed by thermodynamics

Despommier’s book on vertical farming brims with idealism, but is frustratingly short on details. His Columbia faculty colleague, Vincent Racaniello, describes it as a “dream by a particularly good dreamer.” Given Despommier’s admission that he has neither the architecture nor engineering skills to work out the details himself, Racaniello calls for more rigorous analysis from those with appropriate skills.

In her forward to the book, Majora Carter anticipates that future vertical farms will be very different from those in Despommier’s dreams:

In the time between now and the realization of Dickson Despommier’s vision for our food system, there are many opportunities for innovation and entrepreneurship. If the skyscraper farm is like a 747 jetliner, we are now at the stage of the Wright Brothers. […] There will be many failures as a legion of tinkerers and engineers all struggle to take off […]

While it’s possible that Despommier’s visions of vertical farms are akin to the Wright brothers’ early visions of flight, I suspect they’re more like alchemists’ dreams of turning lead into gold. Alchemists didn’t fail for lack of tinkering and engineering. They didn’t even fail because their task was impossible. Today’s physicists can change lead to gold by removing three protons from its nucleus, but doing so requires a vast energy input that is far more costly than the value of the gold produced. Like alchemy, growing food in skyscrapers is technically feasible, but isn’t worth the energy cost.