Industries / Food, farming and forestry / Agriculture and farming / Nursery and horticulture
Watering and feeding
Giving plants in pots, trays, slabs and greenhouse beds the water and nutrients they need, and choosing what their roots grow in: soil, peat, stone wool or a stream of nutrient solution.
7
Milestones achieved, from 1937 to 2024.
2
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
In depth
The same ground is charted in more depth, as an industry of its own.
The road
AchievedWatch forOpen
Achieved
1937
Hydroponics leaves the lab
William F. Gericke of the University of California, Berkeley, adapted water culture, until then a laboratory method for studying plant nutrition, to grow crops for sale with their roots in a nutrient solution instead of soil. He described it in Science on 12 February 1937 under the name "hydroponics". Newspapers then promoted the method, and promoters sold training for a new "profession of soilless farming". The California Agricultural Experiment Station asked D. R. Hoagland and D. I. Arnon for an independent appraisal. Their Circular 347 (December 1938) gave a nutrient recipe and reported that, in one comparison, tomatoes grown in soil in a commercial greenhouse yielded about as much as those in a successful water-culture greenhouse. It also reported that one California firm had stopped using the costly equipment it had installed.
LessonA new growing method still needs the old skills; the circular warned that "without some expert supervision, commercial success is unlikely.".
Credit
William F. Gericke; D. R. Hoagland and D. I. Arnon, University of California
1969
Stone wool replaces soil
Stone wool, also called rockwool, is made by spinning melted rock into fibers. Pressed into slabs and cubes, it gives roots a clean, uniform medium that holds both water and air. Grodan was founded in 1969 and says it invented hydroponic stone wool growing that year. It says its first customers switched to avoid disinfecting greenhouse soil with methyl bromide, a toxic fumigant. A 1988 paper in Acta Horticulturae dates the start in the Netherlands later. It says that by the early 1970s soil diseases and wet, dry or salty soils had stalled yields, and that at the end of the 1970s Grodan introduced rockwool after successful trials around Bleiswijk.
Credit
Grodan (now part of the ROCKWOOL Group); Dutch growers around Bleiswijk
1978
Roots in a film
At the Glasshouse Crops Research Institute in Littlehampton, England, A. J. (Allen) Cooper and colleagues developed the nutrient film technique (NFT). Plants sit in a sloping channel, and a shallow stream of nutrient solution runs past their roots and is pumped round again. In 1978 the team reported that NFT tomatoes gave significantly heavier crops than peat-grown plants with no loss of fruit quality. Lettuce, peppers, chrysanthemums, carnations and strawberries were by then grown commercially this way.
LessonThe authors called precise control of the root zone "essential", with automatic control units built into the system.
Credit
A. J. Cooper, K. Spensley and G. W. Winsor, Glasshouse Crops Research Institute
1990s
Dutch growers must recirculate
In a closed system, the water that drains from the growing slabs is collected, topped up with nutrients and used again instead of running into ground and surface water. Under the Netherlands' Waste Water Disposal Decree, recirculating drain water and collecting rainwater became compulsory for growers using substrates (Van Os, Acta Horticulturae, 1998). Studies showed closed systems saved up to 30% of water and up to 40% of fertilizer compared with open ones (Van Os, Water Science and Technology, 1999).
LessonRecirculation can carry root disease to every plant in a greenhouse, so the solution has to be disinfected. Sodium from poor supply water also builds up until some solution must be let out.
Credit
the Dutch government and greenhouse growers; documented by Erik A. van Os
2013
Sensors schedule nursery watering
From 2010, a US university team installed wireless networks of substrate-moisture sensors in nine commercial nurseries and greenhouses. The networks monitored irrigation and at some sites controlled it, either from moisture set points or from models of each crop's water use. One site was a commercial pot-in-pot nursery, where container plants sit in socket pots sunk into the ground. A cost study there found a 37.5% return on investment and a 2.7-year payback, almost all of it from less time spent managing irrigation.
Credit
John D. Lea-Cox (University of Maryland), Marc W. van Iersel (University of Georgia) and colleagues; cost study by Bruk E. Belayneh, John D. Lea-Cox and Erik Lichtenberg (HortTechnology, December 2013)
2020
Peat-free target missed
Peat is dug from bogs, and it has long been the main ingredient of potting mixes. In 2011 the UK government set voluntary targets for peat-free retail growing media in England by 2020, and for professional growers by 2030. By 2021 peat-free products were 18% of retail growing-media sales, up from 6% in 2011, and peat's share of retail growing media had fallen from 57% to about 30% (Office for the Internal Market, 2023). The government then repeatedly promised a ban on retail peat sales by 2024 but did not legislate one.
LessonVoluntary targets reduced peat use but did not end it. The RHS says the industry still uses 760,000 m³ of peat a year.
Credit
UK government (Defra); figures from the Office for the Internal Market
2024
Peat-free holds its water
The Royal Horticultural Society (RHS) ran 18 months of trials at four UK nurseries (Allensmore, Johnsons, Lovania and Hillier), using growing media from six manufacturers. Under normal nursery watering, five of the six peat-free mixes held similar or more moisture than peat-reduced controls. Every mix produced saleable perennials, shrubs and alpines.
LessonSome peat-free mixes look dry on top while still wet below, so growers may overwater them and wash out nutrients.
Credit
Raghavendra Prasad, RHS Transition to Peat-Free research fellow
Watch for
England's peat sales ban
In October 2025 the UK government committed to "legislate a ban on the sale of peat and peat-containing products when Parliamentary time allows", without a date. In 2026 a petition of about 16,000 signatures from the Peat-free Partnership, whose members include the RHS, The Wildlife Trusts and the National Trust, asked for that law in 2026.
Achieved whena law ending retail peat sales in England is passed and in force, and sales figures show peat products gone from retail shelves.
Claimed by
UK government, October 2025
Learned irrigation cuts water 79%
A decision model that learns when to water each sector of a greenhouse used 79.1% less water than fixed-interval irrigation. It was tested on data from a real multi-sector greenhouse, not by running a live crop.
Achieved whenthe method runs a live greenhouse through a full crop beside grower-run sectors, with metered water and weighed harvest, and a party outside the research team checks the results.
Claimed by
Tommaso Adamo and colleagues, Smart Agricultural Technology, September 2026
Open
01
Tell which pots need water
A camera tool that looks along a mixed bench of potted plants and marks which pots need water now. That is the call a grower makes by eye and by lifting pots. Score the tool against weighed pots and against the grower's own calls.
The briefCredit · unclaimed02
Log nitrate leaving a nursery
A year-long public record of nitrate in the water that leaves one container nursery, at its drainage channel or catch pond. A low-cost open sensor reads it every hour, and a certified laboratory checks it every week.
The briefCredit · unclaimed
Open 01 · the brief
Tell which pots need water
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A camera tool that looks along a mixed bench of potted plants and marks which pots need water now. That is the call a grower makes by eye and by lifting pots. Score the tool against weighed pots and against the grower's own calls.
Why it is still open
Hand watering "requires labor each time plants need water". Seasonal operations still use it because it needs no installed equipment, and as much as 50% of the water from a hose can end up on the floor (George Elliott, University of Connecticut, and Neil Mattson, Cornell University, 2012). The nearest prototype found, a 2025 mobile watering robot, decides by pushing a moisture probe into each pot. A 2024 TU Wien thesis by Tobias Eidelpes built a camera model that rates houseplants as water-stressed or not, with a mean average precision of 0.38 (a detection score where 1.0 is perfect), and does not report checking its labels against pot weight or soil moisture. No camera tool scored against growers' calls on mixed benches was found.
Built with what exists
A Raspberry Pi 5 with its camera module; a low-cost thermal camera such as a FLIR Lepton module, to read leaf and surface temperature; load cells with an HX711 board, or a kitchen scale, for ground truth; Label Studio to label images; an open detection model such as Ultralytics YOLO11.
First step
Weigh and photograph 50 mixed pots on one bench every morning for two weeks, and note which ones the grower waters.
Done looks like
300 pots of at least 10 species followed for 14 days. Publish how often the tool's "water now" calls agree with pot-weight thresholds and with the grower, and how much water the bench would have used under each.
Open 02 · the brief
Log nitrate leaving a nursery
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A year-long public record of nitrate in the water that leaves one container nursery, at its drainage channel or catch pond. A low-cost open sensor reads it every hour, and a certified laboratory checks it every week.
Why it is still open
The open-source, low-cost nitrate monitor of Bulusu and colleagues was validated over five weeks in soil pore water and open water, within 0–10 ppm nitrate-nitrogen (RMSE under 0.2 ppm). A low-cost tree-nursery sensor (Caron and colleagues) was tested in growing media, not in runoff. Searches found no low-cost monitor validated on nursery runoff. The size of the problem is only estimated: a 2013 projection put US savings at about 282,000 kg of nitrogen a year if half of ornamental growers adopted sensor-based irrigation.
Built with what exists
Bulusu and colleagues' published design and building guide for a spectrophotometric nitrate monitor, which measures color after adding a single reagent; a Raspberry Pi or ESP32 data logger; nitrate test strips for daily spot checks; a commercial water laboratory for weekly reference samples.
First step
Take five grab samples of runoff after irrigation at one nursery this month and have a lab test them, to learn the range the sensor must cover.
Done looks like
52 weeks of hourly readings, with each weekly lab result paired with the sensor reading. Report error and drift across the range seen in the runoff, and publish all raw data.