Industries / Food, farming and forestry / Agriculture and farming / Viticulture
Water
Water is the work of deciding when and how much to water vines, and of measuring how thirsty they are, from rain-fed vineyards to drip-irrigated ones in dry regions.
5
Milestones achieved, from 1965 to 2007.
2
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1965
A gauge for vine thirst
P.F. Scholander and colleagues described the pressure chamber, which measures the pull on the water inside a plant by pressing gas on a cut leaf until sap is pushed out of the xylem, the plant's water pipes. In 2001 Xavier Choné and colleagues compared three readings taken this way on vines and found stem water potential the best at telling moderate from severe water shortage. A 2022 paper still calls manual leaf or stem readings with the chamber the "widely established" way to measure plant water status.
Credit
P.F. Scholander, E.D. Bradstreet, E.A. Hemmingsen and H.T. Hammel (Science, 1965); Xavier Choné and colleagues (Annals of Botany, 2001)
1981
Thirst on a schedule
Regulated deficit irrigation (RDI) gives a crop less water than it would use at set stages of growth, to hold back leafy growth without cutting the harvest. A 2022 review dates it to a 1981 paper on peach trees by D.J. Chalmers, P.D. Mitchell and L. van Heek. An FAO report (2002) says RDI raised water use efficiency in fruit trees in Australia's Goulburn Valley by about 60% with no loss of yield, and that work on wine grapes (Goodwin and Jerie, 1992) set the soil-moisture levels that trigger watering by soil type and climate. Australian Grape & Wine lists RDI among the water-saving methods its growers use.
Credit
D.J. Chalmers, P.D. Mitchell and L. van Heek; I. Goodwin and P.H. Jerie for wine grapes
1998
Half-dry roots, then a recount
P.R. Dry and B.R. Loveys proposed partial rootzone drying (PRD): keep part of a vine's roots dry while watering the rest, so the drying roots send chemical signals that slow shoot growth and water use while the crop is kept. In 2009 a meta-analysis (a study that pools the results of many trials) by Victor Sadras found that, against well-watered controls, PRD raised yield per unit of irrigation water by 82% and plain deficit irrigation by 76%, a difference too small to tell apart; when both got the same water, PRD did only 5% better.
LessonMost of PRD's saving came from giving less water, so Sadras concluded it pays only in "rare conditions" given its cost and complexity.
Credit
P.R. Dry and B.R. Loveys (Australian Journal of Grape and Wine Research, 1998); V.O. Sadras (Irrigation Science, 2009)
2006
France lifts its watering ban
A 1964 decree, with orders of 1966 and 1969, banned all irrigation of French vineyards from 1 April to 31 October, according to the agriculture ministry's 2005 answer to a question in the National Assembly. A decree published in December 2006 allowed vines to be watered until 15 August, with stricter rules for appellation wines (wines sold under a protected place name). A further decree in August 2023 moved the last date for watering wine grapes to 15 September.
LessonFrance limited watering to protect the link between appellation wine and its place, and loosened the rules twice as heat and drought threatened harvests; the 2023 change had been promised after a 2022 government consultation on farm water.
Credit
French Ministry of Agriculture; INAO, the appellation authority
2007
Growers buy water in drought
The Murray–Darling Basin holds about 75% to 80% of Australia's irrigated grape production, by Australian Grape & Wine's two counts. In the Millennium Drought its growers had to buy water to keep vines alive: ABARES found 2007–08 and 2008–09 had the highest share of grape growers who were net buyers of water. Australian Grape & Wine says water purchases rose to 21% of growers' total cash costs in the drought, against about 8% on average from 2007–08 to 2014–15, at A$495 to A$702 per megaliter (a million liters).
LessonVines must be watered every year to stay alive, and replanting costs up to A$60,000 a hectare, so growers often keep watering at a loss (Australian Grape & Wine).
Credit
ABARES; Australian Grape & Wine
Watch for
AI irrigation across Prosecco estates
Villa Sandi says weather stations, soil-moisture sensors, satellite and drone images and AI advice across more than 200 ha in northeast Italy cut its water use by about 10% on average. It says plots that also use the University of Milan's drought-tolerant "M" rootstocks use up to 70% less water than neighboring vineyards. It gives no baseline years or metered volumes.
Achieved whenmetered water per hectare and yields, against stated baseline years or matched blocks, are published or checked by an outside party.
Claimed by
Villa Sandi, July 2026
Drones spot thirst early
The EU-funded SmartVitiNet project says its method, which combines drones, hyperspectral cameras (cameras that record many narrow bands of light) and neural networks, detected vine water shortage up to four days before people noticed it, in field tests in Greece and Portugal. The report gives no number of vines and no error rates.
Achieved whena peer-reviewed paper or an outside group reports how often the alerts were right against pressure-chamber readings, across several vineyards and seasons.
Claimed by
SmartVitiNet, reported by the European Innovation Council and SMEs Executive Agency, July 2026
Open
01
Score trunk water sensors
A season-long, side-by-side test of the sensors sold to replace the pressure chamber: a trunk microtensiometer (a chip set into the trunk that reads water potential every few minutes), a trunk dendrometer (which measures the trunk's daily shrinking and swelling) and a sap-flow sensor. All go on the same vines as weekly midday pressure-chamber readings, in at least two vineyards and two varieties, and the result is a public scorecard.
The briefCredit · unclaimed02
Trigger levels for trunk sensors
A table that turns trunk-sensor readings into watering decisions for three wine grape varieties. Log trunk readings beside midday stem readings through a season, fit how the two relate by time of day and air dryness, and then test the resulting triggers on a block the next season.
The briefCredit · unclaimed
Open 01 · the brief
Score trunk water sensors
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A season-long, side-by-side test of the sensors sold to replace the pressure chamber: a trunk microtensiometer (a chip set into the trunk that reads water potential every few minutes), a trunk dendrometer (which measures the trunk's daily shrinking and swelling) and a sap-flow sensor. All go on the same vines as weekly midday pressure-chamber readings, in at least two vineyards and two varieties, and the result is a public scorecard.
Why it is still open
The only independent field test found, by Vinay Pagay at the University of Adelaide, covered one sensor type in two South Australian vineyards. Its trunk readings matched the chamber on mild days but fell below it in the late afternoon on hot, dry days. The only side-by-side comparisons of sensor types found are on a sensor maker's own website.
Built with what exists
PMS Instrument's Pump-Up Chamber for the reference readings; FloraPulse microtensiometers with a METER ZL6 logger; a commercial trunk dendrometer and sap-flow sensor; a weather station for air dryness.
First step
Vines are growing in the southern hemisphere in November. Ask one vineyard there to let you tag ten vines, and start weekly midday stem readings with a borrowed or rented chamber.
Done looks like
One season, two vineyards, two varieties, three sensor types, ten vines per variety. Weekly midday chamber readings and three dawn-to-dusk campaigns. The scorecard publishes each sensor's error against the chamber in megapascals (MPa), the hours when they disagree, uptime and cost, with all raw data released.
Open 02 · the brief
Trigger levels for trunk sensors
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A table that turns trunk-sensor readings into watering decisions for three wine grape varieties. Log trunk readings beside midday stem readings through a season, fit how the two relate by time of day and air dryness, and then test the resulting triggers on a block the next season.
Why it is still open
Pagay wrote that "crop- and cultivar-specific thresholds" of trunk water potential "need to be developed" before the sensors can schedule irrigation where absolute values matter. A Europe PMC search for such thresholds in grapevines found none. In October 2026 the sensor maker FloraPulse described a forecast that predicts when vines "may reach critical stress thresholds", without publishing the thresholds.
Built with what exists
FloraPulse microtensiometers; PMS Instrument's Pump-Up Chamber; a weather station; Python for fitting the curves.
First step
Pick three varieties in one southern-hemisphere vineyard and write the reading schedule: which vines, which hours, and how often.
Done looks like
A published table of trunk trigger values for three varieties, each tied to a stem water potential target. Then a second-season block watered on those triggers, compared row by row with one watered on chamber readings, reporting water used per vine and stem water potential at each irrigation.