Industries / Retail, hospitality and personal services / Retail and e-commerce / Florists
Keeping them cold
Keeping cut stems cold, in clean water and away from ethylene, from the farm's packhouse to the customer's vase.
5
Milestones achieved, from 1938 to 2014.
2
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1938
Flower food in a capsule
In 1938 William Bussert, a rose grower, and James Sykora Sr. of the Chicago wholesaler Amling Co. sold the first FloraLife cut-flower food, a powder in a capsule to which the customer added ordinary sugar. Sugar rationing in the Second World War stalled sales, and the product was relaunched in 1944. In the Netherlands, Camille Buys introduced Chrysal in 1949. Flower foods supply sugar, an acid and a disinfectant, and a sachet now commonly comes with shop-bought flowers.
LessonA product that needs the buyer to add an ingredient stops working when that ingredient runs short.
Credit
William Bussert and James Sykora Sr. (FloraLife); Camille Buys (Chrysal)
1978
Silver blocks ethylene
Ethylene is a gas, given off by ripening fruit, engine exhaust and the flowers themselves, that makes many flowers wilt or drop petals early. In 1978 Veen and van de Geijn reported in the journal Planta that silver bound into a thiosulfate complex (STS) climbs a cut carnation's stem at about 2 m an hour, far faster than plain silver at about 3 cm a day, and still blocks ethylene, so that even a short dip lengthened the flowers' life. By 2001 the Dutch flower auctions required growers to treat lilies, delphiniums, aconitum and other sensitive flowers with STS before sale, and advised it for freesia and gypsophila, according to a market study by the Wageningen institute ATO. The same study said STS has "severe disadvantages" for the environment and workers' health, since silver is a heavy metal.
LessonA treatment that worked well enough to become compulsory left the trade looking for a replacement because of what it is made of.
Credit
Veen and van de Geijn; the Dutch auctions' association (VBN)
2005
Flowers arrive too warm
A white paper by George Staby of the Perishables Research Organization and Michael Reid of the University of California, Davis, discussed with 47 floral companies and posted by the wire service FTD, reported that cut flowers from 58 growers arriving at a California consolidation dock averaged 51 °F (about 11 °C) over a year, against an ideal of 33–35 °F (about 1 °C). At two Florida docks they averaged 50 °F. Earlier work by Celikel and Reid had shown that roses and baby's breath respire about three times faster at 50 °F than at 32 °F, and that 'Ambiance' roses held at 50 °F for five days lasted 7.0 days in the vase, against 11.5 days after five days at 32 °F, 39% less. The paper asked buyers to require a temperature logger or time-temperature indicator in every shipment.
LessonThe cold chain most often breaks at the first step, before the flowers leave the grower.
Credit
George Staby and Michael Reid
2011
Faster beats colder
Ria Leonard, Amy Alexander and Terril Nell of the University of Florida shipped roses, alstroemeria, carnations and gerberas from Bogotá to the United States every month for a year, by a seven-day conventional route with temperature control and by two fast routes, three days and 24 hours, with little or none. On the 24-hour route the flowers sat at or above 10 °C for about 18 hours. Even so, the fast routes gave longer vase life in 83% of rose and alstroemeria shipments, 58% of carnation shipments and 50% of gerbera shipments; roses gained 0.7 to 2.1 days.
LessonDays in the chain cost vase life as surely as warmth does, and the authors still called better temperature control during transport "a critical issue".
Credit
Ria Leonard, Amy Alexander and Terril Nell, University of Florida
2014
Roses cross by sea
In 2014 GreenCHAINge, a group of Dutch flower traders, Wageningen researchers, the Kenya Flower Council and the shipping line Maersk, sent roses in a refrigerated container (reefer) from Mombasa to Antwerp, 25 days in transit. After eight more days of simulated European distribution, their average vase life was ten days, no worse than the same varieties sent by air, and the project said sea transport emitted 87% less CO2. Nine years later, sea freight was still small: a 2023 Wageningen review counted 10 to 20 reefers of Kenyan roses a week to the Netherlands, 2.5% to 5% of the trade, with 10% to 12% of stems lost to Botrytis, a gray mold, which removed most of the cost saving. The Kenya Flower Council aims for half of shipments by sea in 2030.
LessonOne good shipment is not a supply chain; a month of mold risk, and buyers who must order two months ahead instead of days, decide whether sea freight pays.
Credit
the GreenCHAINge partners; Seth Tromp of Wageningen Food & Biobased Research (2023 review)
Watch for
Shop food beats home recipes
FloraLife says its formulated flower foods kept 'Freedom' and 'Vendela' roses and yellow chrysanthemums up to 8 days longer than popular homemade recipes (lemon-lime soda, sugar with vinegar or bleach, a penny in the water), and kept the water clear up to 7 days longer. The test ran in its own laboratory in Walterboro with three vases per treatment. An independent test at North Carolina State University in 2014, by Iftikhar Ahmad and John Dole, on lisianthus, snapdragon, stock and zinnia found that homemade mixes of citric acid and sugar with a disinfectant, and lemon-lime soda, were among the best treatments, while lemon juice and an aluminum sulfate recipe did harm.
Achieved whena laboratory with no stake in flower food repeats the comparison on roses and chrysanthemums with more vases and scorers who do not know the treatments, and publishes the data.
Claimed by
FloraLife, a Smithers-Oasis company, March 2026
Sea freight as strategy
Ruud van der Vliet, a floriculture strategist quoted by the trade magazine THURSD, says shipping roses and chrysanthemums by sea from Kenya, Ecuador, Colombia and Ethiopia is moving from pilot projects to planned strategy, and that a rose arriving after weeks at sea can match one flown in. He puts air freight from Kenya to the Netherlands at about 0.47 kg of CO2 per stem for the airport-to-airport leg and says sea can cut transport emissions by up to about 90%.
Achieved whenan independent count shows a growing share of Kenyan roses reaching Europe by sea, with published rejection rates and vase life equal to air freight.
Claimed by
Ruud van der Vliet, July 2026
Open
01
Log the florist's cooler
A month-long survey of the walk-in and display coolers of 30 florist shops and the vans of 10 delivery drivers, with cheap wireless sensors logging temperature and humidity every five minutes, published as an open dataset beside the vase life of a reference rose bought from each shop.
The briefCredit · unclaimed02
Logger trace to vase days
A free tool that reads the temperature log from a flower box or cooler and estimates how many days of vase life the stems have lost, tested against roses whose real vase life was measured.
The briefCredit · unclaimed
Open 01 · the brief
Log the florist's cooler
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A month-long survey of the walk-in and display coolers of 30 florist shops and the vans of 10 delivery drivers, with cheap wireless sensors logging temperature and humidity every five minutes, published as an open dataset beside the vase life of a reference rose bought from each shop.
Why it is still open
The measured surveys found cover the docks upstream: the 2005 white paper logged flowers arriving at consolidation docks, not the shop. University of Florida trials by Nell and Leonard showed in florist display coolers that lilies kept at 35 °F lasted twice as long as at 70 °F, and warned that shops trust the thermostat setting and hang thermometers where the door lets in warm air. A search for a published, logged survey of florists' own coolers found guidance articles and one supplier's warehouse, but no shop survey.
Built with what exists
Xiaomi LYWSD03MMC Bluetooth thermometers flashed with the open pvvx firmware, or USB loggers such as the Elitech RC-5; a probe in a small vial of water to stand in for the temperature of the stems; Home Assistant or a Raspberry Pi to collect readings; a vase-life room held at 20 °C with 12 hours of light a day.
First step
Buy ten sensors, flash the firmware, check them against an ice bath, and ask five local florists to host one each.
Done looks like
Four weeks of five-minute readings from 30 coolers and 10 vans, published with the share of time spent between 1 °C and 4 °C (34–39 °F), the share above 8 °C, door-opening spikes and humidity; and the vase life of one rose cultivar bought weekly from five of the shops, set against each shop's record.
Open 02 · the brief
Logger trace to vase days
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A free tool that reads the temperature log from a flower box or cooler and estimates how many days of vase life the stems have lost, tested against roses whose real vase life was measured.
Why it is still open
The models exist. Tromp and colleagues at Wageningen showed in 2012 that a simple time-temperature sum predicts lost rose vase life well between 2 °C and 6 °C but fails below 2 °C on long storage and underestimates vase life above 6 °C. Van Meeteren and colleagues (2015) found a curved, sigmoidal model fitted far better, and wrote that techniques to measure the potential vase life at the point of sale "are not available at this moment". The 2023 Wageningen sea-freight review still asked whether 30 days at 0.5 °C and two days at 7–8 °C, the same sum of 15 degree-days, leave the same vase life. A 2016 master's thesis at Delft University of Technology, written for KLM Cargo by F.C.T. van der Voort, applied a degree-days model to the logs of 30 shipments of Kenyan roses flown from 2011 to 2015 and estimated about one day of vase life lost in transit, but it did not measure the roses' vase life to check the estimate. A search found no public tool that applies these models to a logger file, or any test of them on real deliveries to florists.
research.wur.nlresearch.wur.nledepot.wur.nlrepository.tudelft.nl
Built with what exists
USB temperature loggers such as the Elitech RC-5; two household refrigerators held at set temperatures with Inkbird ITC-308 controllers; Python with SciPy to fit the time-temperature-sum and sigmoidal models; rose stems from a wholesaler; a web page that accepts a logger's CSV file.
First step
Code the 2012 time-temperature-sum model from the paper, run it on the paper's own storage conditions, and check that it reproduces the published results.
Done looks like
300 stems of two rose cultivars put through ten logged temperature histories (steady and changing, 1 °C to 12 °C, 2 to 20 days), with the tool's predicted vase life against the measured vase life and a published average error of 1.5 days or less; then 20 real deliveries to florists logged and vase-tested the same way.