Industries / Food, farming and forestry / Agriculture and farming / Produce packing
Washing and food safety
Cleaning produce and stopping germs from spreading from one fruit to the next, through the wash water, the equipment, the building, and the standards, rules and audits that check them.
7
Milestones achieved, from 1997 to 2020.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1997
Retailers write a farm standard
In 1997 a group of European retailers set up EurepGAP to write one certification standard for good agricultural practice (G.A.P.) in place of each retailer's own. The first producer certificates followed in 2001, the first group certificate going to the Fyffes Group's associated banana growers in Belize and Costa Rica. Renamed GLOBALG.A.P. in 2007, its fruit and vegetable standard runs from preharvest work to postharvest handling, and the body says its standards are applied in more than 130 countries. A study of Kenya's export vegetable growers counted about 11,100 smallholders in nine districts in late 2005, of whom about 350 were certified.
LessonA standard written by buyers became a gate that most small growers had not yet passed.
Credit
the EurepGAP retailers; FoodPLUS GmbH, owner of the standard; Dagmar Mithöfer, Eddah Nang'ole and Solomon Asfaw for the Kenya count
1998
Wash water warmer than fruit
In October 1998 the US Food and Drug Administration (FDA) and USDA published a guide for growers, packers and shippers of fresh fruit and vegetables. It was guidance, not law. It advised packers to set water-change schedules, to clean and sanitize dump tanks, flumes, wash tanks and hydrocoolers, and, for apples, celery and tomatoes, to keep wash water warmer than the produce: colder water draws water, and any germs in it, into the fruit, where washing cannot reach. It cited work by Zhuang, Beuchat and Angulo (1995), in which tomatoes at 25 °C dipped into a 10 °C bacterial suspension took up more Salmonella into their core tissue than tomatoes dipped at 25 or 37 °C.
LessonA wash can push germs in as well as rinse them off.
Credit
FDA and USDA; R.-Y. Zhuang, L. R. Beuchat and F. J. Angulo
2006
One plant, 26 states
In August and September 2006, E. coli O157:H7 in bagged spinach made 199 people ill in 26 US states, by CDC's count of 6 October 2006. Of these, 102 were hospitalized, 31 developed a kind of kidney failure (hemolytic-uremic syndrome) and three died. The outbreak strain was found in 13 packages, and 11 of them carried lot codes from a single processing plant on one day. The products named in the outbreak were brands processed by Natural Selection Foods. In 2007 California's leafy greens shippers set up the Leafy Greens Marketing Agreement (LGMA), whose members accept mandatory audits by government inspectors; the LGMA says its members now produce about 90% of US leafy greens.
LessonWhen one plant washes and bags greens for many brands, one contaminated lot reaches all of them.
Credit
CDC and state health departments; the California leafy greens industry
2011
Chlorine must not run out
Free chlorine is the active part of the chlorine added to wash water, and dirt and plant juice use it up. Washing cut lettuce, Yaguang Luo and colleagues at the USDA Agricultural Research Service found that 1 mg/L of free chlorine before washing was quickly spent and let E. coli O157:H7 survive and spread to clean leaves; at 5 mg/L a few cells still spread in four of seven trials; at 10 mg/L or more no spread was detected. Rewashing contaminated lettuce in 50 mg/L chlorine for 30 seconds did not remove the bacteria. In a 2012 pilot-plant study the same group saw spread "frequently" whenever free chlorine dropped below 1 mg/L.
LessonWashing cannot clean up afterwards; its job is to stop germs spreading, so the sanitizer has to be kept up all the time.
Credit
Yaguang Luo and colleagues, USDA ARS, Beltsville, Maryland
2011
A clean audit, deadly shed
Listeria from cantaloupes packed at Jensen Farms in Colorado infected 147 people in 28 states and killed 33, by CDC's count. FDA called it the first listeriosis outbreak linked to a whole, unprocessed fruit or vegetable. Thirteen of 39 swabs from the packing shed carried outbreak strains. FDA named as likely factors the washing and drying equipment, bought used in July 2011 after earlier use on potatoes, which could not be cleaned properly, water pooling on the floor, and the lack of any cooling step to remove field heat before cold storage. A US House committee later found that a private auditor, Bio Food Safety, working under Primus Labs, had given the shed a "superior" 96% score in July 2011 while noting that no antimicrobial was used in the wash water.
LessonAn audit that scores against a basic checklist can pass a shed that the regulator's guidance would fail.
Credit
CDC; FDA with Colorado state and county health officials; House Energy and Commerce Committee
2016
Packing comes under federal rules
FDA's Produce Safety Rule took effect on 26 January 2016. By FDA's account it set, for the first time, minimum standards for growing, harvesting, packing and holding fruit and vegetables in the United States, with the largest farms due to comply from January 2018. Water used during and after harvest must have no detectable generic E. coli in 100 mL and may not be untreated surface water, and farms must keep water-change schedules, watch for build-up of soil and plant debris, and keep water at a temperature that limits infiltration into the produce. FDA stopped holding off enforcement of these water rules from January 2023 for the largest farms. Packhouses that do work outside the rule's definition of a farm come under FDA's preventive controls rule for food facilities instead.
Credit
FDA, under the Food Safety Modernization Act
2020
Europe caps chlorate residues
Chlorate is a by-product of chlorine disinfectants used to treat drinking water and food-processing water, and it can stop the thyroid taking up iodine. Monitoring from 2014 to 2018 found it in food above the EU's default limit of 0.01 mg/kg. Commission Regulation (EU) 2020/749, applicable from 28 June 2020, set temporary limits instead, such as 0.7 mg/kg for leaf vegetables and herbs, 0.1 mg/kg for tomatoes and 0.05 mg/kg for apples, citrus and stone fruit, and kept foods for infants and young children at 0.01 mg/kg.
LessonThe sanitizer that keeps wash water safe leaves its own residue, so packers have to dose to a measured level, not simply add more.
Credit
European Commission; European Food Safety Authority (EFSA)
Watch for
Germ counts from the wash
Drizzle Health says its MagnaFlow analyser connects to existing wash tanks and measures pathogens and spoilage organisms, including E. coli, Salmonella and common fungi, as produce moves through the line, without the enrichment step that adds 18 to 48 hours to standard tests. It is recruiting produce processors for a free pilot.
Achieved whenan independent laboratory runs it beside standard culture methods on commercial wash water and publishes its detection limits and how often the two agree.
Claimed by
Drizzle Health, February 2026
Open
01
Which crops drink their wash
A public table, with its method and photos, of how much water each common fresh crop takes in when dipped in water colder than itself, measured with a food dye instead of germs: for each crop, how far the dye spreads inside the fruit at several temperature gaps, depths and dipping times.
The briefCredit · unclaimed02
Is the fruit dry yet
A camera check at the end of a packing line's dryer that flags fruit still wet from washing, using a thermal camera (wet skin reads cooler because its water is evaporating) beside an ordinary camera, with an open dataset of fruit labelled by the water left on them.
The briefCredit · unclaimed
Open 01 · the brief
Which crops drink their wash
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public table, with its method and photos, of how much water each common fresh crop takes in when dipped in water colder than itself, measured with a food dye instead of germs: for each crop, how far the dye spreads inside the fruit at several temperature gaps, depths and dipping times.
Why it is still open
FDA's rule asks farms to keep post-harvest water at "a temperature that is appropriate for the commodity" to limit infiltration, but its 1998 guide named only apples, celery and tomatoes as susceptible and carrots as unaffected. Studies found in a Crossref search in November 2026 concentrate on tomatoes (Journal of Food Protection, 2012 and 2016), and no table across many crops was found.
Built with what exists
FD&C Blue No. 1 (Brilliant Blue FCF), a permitted food colour; temperature-controlled water baths; a phone camera on a copy stand or a flatbed scanner for cut sections; Fiji (ImageJ) to measure the dyed area; Python for the statistics.
First step
Dip 20 tomatoes, which the literature says take up water, and 20 carrots, which FDA's guide says do not, in dyed water 15 °C colder than the produce, to check that the method tells the two apart.
Done looks like
At least 12 crops, each tested with water 0, 5 and 15 °C colder than the fruit, at two depths and two dipping times, 20 fruit per condition, with a dyed-area figure for every fruit and all photos published; the tomato results set against the published Salmonella internalization studies.
Open 02 · the brief
Is the fruit dry yet
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A camera check at the end of a packing line's dryer that flags fruit still wet from washing, using a thermal camera (wet skin reads cooler because its water is evaporating) beside an ordinary camera, with an open dataset of fruit labelled by the water left on them.
Why it is still open
FDA's review of the 2011 Jensen Farms outbreak named free moisture on washed cantaloupe rinds, and condensation on warm fruit put into cold storage, as conditions that helped Listeria grow. Crossref searches in November 2026 for detecting surface water on produce after washing found no published method.
Built with what exists
A low-cost radiometric thermal camera core such as the Teledyne FLIR Lepton 3.5 (160 by 120 pixels); a Raspberry Pi 5 with its camera module; a short lab conveyor or turntable; a precision balance to weigh the surface water (each fruit weighed before and after blotting); scikit-learn or a small neural network.
First step
Wash and partly dry 50 tomatoes, weigh each to find its surface water, and take a thermal image of each to see whether the temperature drop tracks the water left on it.
Done looks like
For three crops, for example cantaloupes, tomatoes and apples, 1,000 fruit each labelled by measured surface water, and a detector that finds fruit carrying more than a set amount of water with at least 95% sensitivity at a 5% false-alarm rate on fruit it was not trained on; dataset and code published.