Industries / Food, farming and forestry / Agriculture and farming / Nursery and horticulture / Vertical farming and hydroponics
Climate and water
Keeping the air around the plants at the right temperature, humidity and carbon dioxide level, keeping the water around the roots clean, and keeping the crop safe to eat.
6
Milestones achieved, from 1980s to 2021.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1980s
Strawberries fail in nutrient film
The nutrient film technique (NFT) grows plants with their roots in a shallow stream of nutrient solution and no growing medium. Strawberry growers in Europe adopted it in the 1970s and had largely abandoned it by the 1980s "following systematic failures", according to a 2026 review by Shalyne Scott and Camilo Villouta. They trace the failures to two linked causes. Over a six-month fruiting season the root mat thickens and starves the stream of oxygen, and roots short of oxygen fall to Pythium, a water mold that rots roots.
LessonA root system proven on one crop is not proven on another; the review says oxygen limits set for tomato and cucumber do not apply to strawberry.
Credit
European strawberry growers (the failure); reviewed by Shalyne Scott and Camilo Villouta
1992
Air blown into lettuce hearts
Lettuce grown fast indoors often gets tipburn: brown, dead edges on the young inner leaves, which lose too little water to draw in enough calcium. E. Goto and T. Takakura grew butterhead lettuce in nutrient solution in a closed growth room and blew room air onto the inner leaves through pipes. Untreated plants showed tipburn at about 15 g fresh weight; with air supplied day and night, none showed it up to 60 g (Transactions of the ASAE, 1992). Others have since confirmed the cause and the cure. John Ertle and Chieri Kubota found visible tipburn in all 10 commercial cultivars they grew under indoor-farm conditions (2023), and Parker Persons and colleagues found that stronger downward airflow, up to 1.3 m/s, cut tipburn without reducing growth (2025).
LessonMore light means faster growth and more tipburn, so a farm that adds light must also add air (Persons and colleagues).
Credit
E. Goto and T. Takakura
1995
Root water disinfected before reuse
Growing without soil did not end root disease. Pythium, Phytophthora and some plant viruses spread easily in water that is collected and used again. W. Th. Runia reported in 1995 that heat treatment, ozone or ultraviolet (UV) light disinfected recirculated water on more than 500 Dutch nurseries. Heat and ozone were "hardly feasible" on nurseries smaller than 1 ha, so Erik van Os and colleagues tested slow sand filters, beds of sand through which water seeps slowly while microbes in the sand trap pathogens. Fine and medium sand at 0.1 m³ per m² per hour removed Phytophthora cinnamomi completely, but Fusarium and tomato mosaic virus were only delayed (1998). In W. Wohanka's pilot filters at Geisenheim, Germany, a Fusarium load of about 79,000 spores per milliliter came out at no more than 88, but the filtered water lost most or all of its oxygen (1995).
LessonA cheap filter is selective: it stops some pathogens and only slows others, so a grower has to know which disease it is guarding against.
Credit
W. Th. Runia, E. A. van Os and colleagues (Netherlands); W. Wohanka (Germany)
2004
A greenhouse with shut windows
From 1998 to 2003 the Dutch engineering firm Ecofys built and tested a greenhouse whose vents never open. A heat pump, an underground aquifer that stores summer heat for winter, and air ducts under the crop control temperature, humidity and carbon dioxide. In a 1,400 m² demonstration greenhouse Ecofys reported 20 to 35% less fossil fuel, 20% more tomatoes, 80% less chemical crop protection and 50% less irrigation water, and the first 14,000 m² went in at a commercial grower. Wageningen researchers measured 56.2 kg/m² of tomatoes in a closed greenhouse in 2002, 22% more than a crop model predicted for a conventional greenhouse, and found that its constant 1,000 ppm of carbon dioxide explained only 9 points of the gain. The results were presented at the GreenSys 2004 conference.
LessonA climate system can be taken up in pieces. "Next Generation Growing", with insulating screens and forced ventilation but no aquifer, grew 69 kg/m² of tomatoes on 750 MJ/m² of heat in a 1,000 m² trial and was in commercial use by 2012 (de Gelder and Dieleman).
Credit
J. J. G. Opdam and colleagues, Ecofys; A. de Gelder and E. Heuvelink, Wageningen
2005
Refinery CO2 piped to greenhouses
Crops grow faster in air with more carbon dioxide (CO2) than outdoor air holds, so greenhouse growers add it. Dutch growers mostly got it by burning natural gas in their heating boilers, which in summer wasted the heat, or bought liquid CO2 delivered by tanker. In 2005 OCAP began piping CO2 captured at Shell's Pernis refinery near Rotterdam to greenhouses, through a pipeline taken over from the government, and a bioethanol plant became a second source in 2011, by OCAP's account. A 2022 report by the Netherlands Environmental Assessment Agency (PBL) and TNO, citing Linde Gas, puts OCAP's deliveries at roughly 400,000 t of CO2 a year to greenhouses in the Rotterdam and Amsterdam regions, from Shell Pernis and the Alco Energy bioethanol plant, and the CO2 that burning natural gas for it would have released at 205,000 t.
LessonWith a single source every outage reached the growers at once, and demand outran supply until a second source was found, in OCAP's own account.
Credit
OCAP (now part of Linde Gas Benelux); Shell Pernis refinery; Alco Energy Rotterdam
2021
Salmonella in indoor-grown greens
In 2021, 31 people in Illinois, Wisconsin, Pennsylvania and Michigan fell ill with Salmonella Typhimurium, and 4 were hospitalized. Investigators traced the illnesses to packaged salad greens from BrightFarms' greenhouse in Rochelle, Illinois, which grew them on rafts floating in ponds of nutrient solution. FDA calls it what it believes was its first US investigation of an outbreak tied to leafy greens from controlled environment agriculture (CEA), the trade's term for growing crops indoors or under cover. The pond water was sand-filtered and UV-treated on the way in but not routinely disinfected once in the ponds. FDA found a different Salmonella in one production pond, the outbreak strain in a stormwater basin about 25 feet (8 m) from the building, chiller pipes dripping condensation onto produce, and a harvest blade that could not reliably leave out leaves that had touched pond water.
LessonGrowing indoors does not keep pathogens out. FDA did not find how the bacteria got in, and pointed to pond water, outdoor storage of growing media and the land next door.
Credit
US Food and Drug Administration and Centers for Disease Control and Prevention (investigation); BrightFarms (grower)
Watch for
Additive in place of fans
Vertical airflow fans stop tipburn but cost energy, shade the crop and are hard to fit. In a summer greenhouse trial with lettuce grown on rafts, a calcium-mobilizing biostimulant added to the nutrient solution at 0.5 mL per liter cut tipburn by 71 to 75% at harvest (28 days) without fans, with no loss of final yield, which the authors call comparable to fans.
Achieved whena farm or research group other than the authors repeats it in an indoor vertical farm over several crop cycles and reports saleable heads with the additive against heads under fans.
Claimed by
Moein Moosavi-Nezhad and Qingwu Meng, Frontiers in Plant Science, November 2025
Open
01
Tag recalls by growing method
A public table of every recall, outbreak notice and warning letter since 2010 that involves produce grown in greenhouses, vertical farms, or hydroponic or aquaponic systems, in the US, Canada, the UK and the EU. Each record is tagged with growing method, crop, pathogen or hazard, and the cause the investigators named, with a link to the original.
The briefCredit · unclaimed02
Oxygen limit for strawberry roots
A growing trial that finds the lowest dissolved oxygen, measured along the channel, at which strawberry roots in NFT stay healthy through a full fruiting season, and that measures how much the growing root mat slows the flow.
The briefCredit · unclaimed
Open 01 · the brief
Tag recalls by growing method
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public table of every recall, outbreak notice and warning letter since 2010 that involves produce grown in greenhouses, vertical farms, or hydroponic or aquaponic systems, in the US, Canada, the UK and the EU. Each record is tagged with growing method, crop, pathogen or hazard, and the cause the investigators named, with a link to the original.
Why it is still open
FDA's 2022 report calls its 2021 BrightFarms investigation what it believes was its first US investigation of an outbreak from indoor-grown leafy greens, so the record is short and scattered. openFDA's food recall records have no field for growing method: a search of product descriptions for "hydroponic" returns three records (Hilltop Farm microgreens, 2015; Kalera baby romaine, 2022), and the method appears only when a description happens to mention it. Two 2025 reviews (Lewis Ivey and colleagues; Cheruiyot and Mechesso) summarize lab and farm studies of pathogens in soilless systems, not recalls by growing method, and GitHub's recall scrapers do not tag growing method.
Built with what exists
the openFDA food enforcement API; CDC outbreak notices; the Canadian Food Inspection Agency's recall list; the UK Food Standards Agency's alerts; the EU's RASFF Window; Google's Gemini Decision 1, which scores yes/no and rubric questions for up to 1,000 items in one pass, to suggest tags (see the Note "A thousand questions, one pass"), with every tag checked by a person.
First step
Pull every leafy-green, herb and microgreen record from the openFDA food enforcement API and tag the first 200 by hand to set the rules.
Done looks like
every produce record from the four regions, 2010 to 2026, screened; each indoor-grown record tagged; a random 10% re-tagged by a second person with at least 95% agreement; the table and code published and updated every quarter.
Open 02 · the brief
Oxygen limit for strawberry roots
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A growing trial that finds the lowest dissolved oxygen, measured along the channel, at which strawberry roots in NFT stay healthy through a full fruiting season, and that measures how much the growing root mat slows the flow.
Why it is still open
Scott and Villouta (May 2026) name two gaps that must be closed before NFT can be reconsidered for strawberries: how the root mat changes the channel's flow, and "a strawberry-specific dissolved oxygen threshold under NFT-relevant conditions". The nearest trials found are short or run with a supplier. In a 2020 trial by the Dutch research firm Delphy with Moleaer, a maker of nanobubble equipment, strawberries given water at 30 mg/L of oxygen yielded 8.2 kg against 7.2 kg at 7 to 9 mg/L, in a growing system the report does not describe. An Arizona State University trial in deep water culture found no difference in growth after four weeks between about 70% and 100% oxygen saturation, with fruit results still to come (ASHS 2024 abstract). GitHub searches for NFT strawberry and hydroponic oxygen logging found nothing.
Built with what exists
NFT channels with an air pump or oxygen injection to set levels; optical dissolved-oxygen probes, such as Atlas Scientific's, at both ends of each channel; flow meters; a Raspberry Pi to log readings; everbearing strawberry plants under LED grow lights; a camera for weekly root photos.
First step
Build two 3 m NFT channels with oxygen probes at both ends, plant ten young strawberry plants, and log how oxygen falls along the channel for a month.
Done looks like
one six-month fruiting cycle at three or more oxygen levels with two channels each, publishing hourly oxygen and flow logs, fruit weight per plant and weekly root-browning scores, and the oxygen level below which yield or root health falls.