Industries / Food, farming and forestry / Agriculture and farming / Nursery and horticulture / Vertical farming and hydroponics
Lighting
Giving plants electric light, either as their only light in a closed room or on top of sunlight in a greenhouse, and choosing its color, strength and timing.
7
Milestones achieved, from 1880 to 2024.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1880
Melons grown under arc light
The engineer C. William Siemens set a 1,400-candle electric arc lamp about 2 m above the glass of a sunken melon house at Sherwood and lit pots of mustard, carrots, swedes, beans, cucumbers and melons from 5 to 11 each evening. Plants given both daylight and arc light grew more vigorously and darker green than plants given either alone, though he judged daylight about twice as effective as the arc. In a second note that month he reported strawberries ripening faster when the lamp ran through the night. Both reports were read to the Royal Society in March 1880.
LessonSiemens blamed much of the shortfall on thick condensation on the glass, which "greatly obstructed the action of the light"; light lost between lamp and leaf is paid for and wasted.
Credit
C. William Siemens
1920
Day length controls flowering
W. W. Garner and H. A. Allard of the US Department of Agriculture found that Maryland Mammoth tobacco and other plants flowered or kept growing leaves depending on the length of the day, not on how bright the light was. They tested it by wheeling plants into a dark house to shorten their days, and coined the words photoperiod and photoperiodism. Their review notes earlier growers who added electric arc light at night and made spinach flower and lettuce grow faster. Lamp schedules for crops are still set by the daily lighting period and the daily light integral, the total light a plant receives in a day (Moe and colleagues, 2006).
Credit
W. W. Garner and H. A. Allard, USDA Bureau of Plant Industry
1970s
Sodium lamps light greenhouses
Norwegian greenhouses took up fluorescent lamps from 1950, mainly for raising young plants, and used them until about 1980. High-pressure sodium (HPS) lamps arrived in the 1970s and turned electricity into photosynthetically active radiation (PAR, the 400–700 nm light plants use for growth) more efficiently than fluorescent or metal halide lamps, according to a 2006 review by Moe, Grimstad and Gislerød. The review treats lamps as the basis of year-round greenhouse production in Norway's dark winters.
Credit
lamp makers and Norwegian growers; recorded by Moe, Grimstad and Gislerød
1994
Bright blue LEDs arrive
Red and green light-emitting diodes (LEDs) had existed for decades, but a bright blue one had resisted researchers for three decades, the Nobel committee wrote. Isamu Akasaki and Hiroshi Amano at Nagoya and Shuji Nakamura at Nichia Chemical Industries produced bright blue beams in the early 1990s, and Nakamura's team reported candela-class indium gallium nitride blue LEDs in Applied Physics Letters in March 1994. White LEDs, now common in grow lights, are blue LEDs coated with a phosphor that turns part of the blue into longer wavelengths, and the best blue LEDs convert 93% of their electricity into light, the highest of any color (Kusuma, Pattison and Bugbee, 2020).
Credit
Isamu Akasaki, Hiroshi Amano and Shuji Nakamura (Nobel Prize in Physics, 2014)
2020
Far-red light counts after all
The standard definition of PAR stops at 700 nm, which leaves out far-red light (700–750 nm). Shuyang Zhen and Bruce Bugbee of Utah State University measured 14 crop species and found that adding far-red photons, up to 40% of the total, raised whole-canopy photosynthesis as much as adding the same number of 400–700 nm photons, and proposed widening PAR to 750 nm. Working separately at Wageningen University & Research, Jin, Urbina, Heuvelink and Marcelis added 52 µmol m⁻² s⁻¹ of far-red to red-blue LED light and raised lettuce dry weight by 46–77% after four weeks, mainly because bigger leaves caught more light.
LessonFar-red makes lettuce leaves spread but makes the stems of many other species stretch, so it is not a free gain for every crop (Kusuma and colleagues).
Credit
S. Zhen and B. Bugbee; W. Jin, J. L. Urbina, E. Heuvelink and L. F. M. Marcelis
2023
Neighbors win darker nights
Lamps run at night in glasshouses light the sky above them. Leamington, Ontario, passed a light-abatement bylaw in June 2022 that required growers who light their greenhouses to close curtains from 5 p.m. to 8 a.m. from October 2022. In February 2023 the municipality laid 88 charges against 12 greenhouse companies, CBC News reported, while the mayor said most greenhouses complied.
LessonClosing light curtains also blocks most ventilation and makes humidity and temperature harder to control, says David Lubitz of the University of Guelph, who began measuring escaping light because "there's almost no measurements".
Credit
Municipality of Leamington; reported by CBC News
2024
LEDs replace sodium lamps
Wageningen Social & Economic Research's Energiemonitor counts energy use in Dutch greenhouses each year. For 2024 it reports that growers bought 2.8 billion kWh of electricity from the grid, 14% less than in 2023, because LEDs replaced high-pressure sodium lamps, lighting was used more selectively, and growers used more power from their own gas engines. Total electricity use was about 6.4 billion kWh, an average of 64 kWh per m² of greenhouse, slightly up as growers lit more again after the energy-price crisis.
Credit
Dutch greenhouse growers; measured by Pepijn Smit, Wageningen Social & Economic Research (report 2025-150)
Watch for
Full-LED tomatoes, half the heat
Signify, which makes Philips grow lights, and Grodan, which makes stone wool, report a three-year trial in which tomatoes under LED top lights, with insulating screens and active air dehumidification, needed more than 50% less heat from the heating pipes than commercial practice. In the third year (week 39 of 2025 to week 16 of 2026), adding LED lights between the plant rows raised truss tomato yield by 10% over top lights alone, and the LEDs used 235.5 kWh per m², 68.8% of the trial's energy. The companies say running both light setups in one compartment limited the comparison, and the heat saving is measured against "practice", not a parallel compartment.
Achieved whena grower or research station with no stake in the products runs a full season with metered heat and electricity per kilogram of tomatoes, beside a same-size compartment grown the usual way.
Claimed by
Signify and Grodan, June 2026
Open
01
Measure aged grow lights
A public tally of how much light horticultural LED fixtures still give after years in service, measured on site at night with a calibrated quantum sensor (a meter that counts plant-usable photons) and compared with each maker's datasheet and lifetime claim.
The briefCredit · unclaimed02
Map greenhouse lamps from orbit
A map of the lit greenhouses in one growing region, such as Westland in the Netherlands or Leamington–Kingsville in Ontario, that tells LED-lit glasshouses from sodium-lit ones by the color of their night glow, and tracks month by month how much light each one lets escape before and after local curtain rules.
The briefCredit · unclaimed
Open 01 · the brief
Measure aged grow lights
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public tally of how much light horticultural LED fixtures still give after years in service, measured on site at night with a calibrated quantum sensor (a meter that counts plant-usable photons) and compared with each maker's datasheet and lifetime claim.
Why it is still open
Lifetime claims such as L70 or Q90 (hours until output falls to 70% or 90%) are projected from lab tests of bare LED packages (IES LM-80 and TM-21). Kusuma, Pattison and Bugbee write that these projections leave out losses in the fixture's optics, faster aging at higher temperatures and driver failures, and that many fixtures claim lifetimes beyond the six-fold projection the method allows. Web and Crossref searches found only package-test standards, a Utah State University check of used sodium lamps, and makers' own endurance tests and anecdotes, with no published multi-year field measurements of LED grow lights.
Built with what exists
A full-spectrum quantum sensor such as an Apogee SQ-500 series, or a handheld spectroradiometer to catch color shift as phosphors age; a fixed jig that holds the sensor at a set distance below a fixture; a laptop or Raspberry Pi to log readings; the farm's purchase records for each fixture's age and run hours.
First step
Ask one greenhouse or indoor farm that bought fixtures in batches over several years to let you measure ten fixtures from each batch at night, after the lights have warmed up.
Done looks like
At least 100 fixtures of at least three makes and three age groups (under 1 year, 3–4 years, 6 years or more) measured on a nine-point grid at fixed distance, with output as a percentage of a new fixture of the same model and the maker's claimed lifetime beside it. Readings, photos and method are published.
Open 02 · the brief
Map greenhouse lamps from orbit
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A map of the lit greenhouses in one growing region, such as Westland in the Netherlands or Leamington–Kingsville in Ontario, that tells LED-lit glasshouses from sodium-lit ones by the color of their night glow, and tracks month by month how much light each one lets escape before and after local curtain rules.
Why it is still open
The Energiemonitor reports LEDs replacing sodium lamps but gives no LED share of the lit area, and Leamington enforces its curtain rule by inspection and charges laid case by case. DarkSky International notes that a single large greenhouse can outshine its host town. The nearest work found: University of Guelph engineers fly drones over single greenhouses, DarkSky follows the region's total night glow, and Geist and colleagues (2026) used multispectral night images from the SDGSAT-1 satellite to see light fall in color and strength in US cities with lighting rules. No greenhouse-by-greenhouse map of lamp type or curtain use was found.
Built with what exists
SDGSAT-1 multispectral night images, as used by Geist and colleagues; monthly VIIRS night-light composites from NASA's Black Marble product; greenhouse outlines from OpenStreetMap; Python with rasterio and geopandas; a handheld spectrometer read from public roads at night, and growers' lamp types reported in the trade press, as ground truth.
First step
Register with the SDGSAT-1 Open Science Program, whose account checks take several days, and meanwhile overlay OpenStreetMap greenhouse outlines on VIIRS monthly composites over Leamington to see whether single greenhouses can be told apart.
Done looks like
Every lit greenhouse over 1 ha in the region labeled LED, sodium or mixed, with accuracy reported against at least 50 greenhouses whose lamp type is known, plus a monthly series of escaped light per greenhouse from 2022 to 2026. Map, code and labels are published.