For decades, one number has been circulating: 140 liters of water per cup of coffee. It gets copied, cited, and written into law — rarely questioned. The picture it paints of coffee is of a water-guzzling crop. That picture is wrong. Some of our producers come close to zero in actual consumption. Water problems in coffee production are real — just not the ones being discussed.
Whether coffee production is environmentally sound gets raised regularly, sometimes in sensationalist terms. Too much water use, too much monoculture, too many inputs. Some of that is true, but there are many counterexamples showing how coffee can be produced differently. Just as diverse as the brands and names coffee roasters invent, so diverse is how coffee is grown. In full sun, in full shade. In monoculture, in polyculture. Wet-processed, dry-processed, or somewhere between.
That these different approaches use different amounts of water is obvious. Yet the figure of 140 liters per cup persists and keeps getting repeated without scrutiny.
So I want to take a closer look at those 140 liters and understand the calculation behind them. Our coffee partners shared their own numbers with me — and to give away the punchline: the 140 liters are wrong. The actual figure is as individual as coffee itself.
Where does the number come from?
The often-cited figure of 140 liters of water per cup is technically accurate but deeply misleading. Around 96% of those 140 liters is rainwater that falls on the coffee plant and evaporates. That's water which would have fallen on the same land and evaporated through natural vegetation regardless of whether coffee was growing there. According to Revolve, only about 1.4 liters per cup actually come from irrigation (ground or surface water). Around four liters represent the theoretical dilution requirement for fertilizer runoff.
I'm still struck by how uncritically those 140 liters get copied without anyone questioning them. It's not much more than a headline number. So it's no surprise to find that the figure comes from a 2003 calculation based on climate data from national capital weather stations, with no regional differentiation, treating rainwater and irrigation water as equivalent.
In 2003, Dutch researchers Hoekstra and Chapagain, in their widely cited paper "the water needed to have the Dutch drink coffee," calculated those 140 liters per cup. They drew on FAO climate data (CLIMWAT). There's a catch, as they themselves acknowledge: they used climate data from the capital cities of coffee-producing countries, not from actual coffee-growing regions.
From my own years of experience: comparing the climate data from Managua, the capital of Nicaragua, with that of Finca Santa Rita in northern Nicaragua reveals a huge difference. Managua runs about 10 degrees hotter on average than Santa Rita, has no tree cover, and is considerably drier.
We admit that this is a crude assumption, because the climate near the capital is not necessarily representative for the climate in the areas in the country where coffee is grown, but global data on exact locations of coffee plantations are not easily obtainable.
Getting precise coordinates for coffee farms in 2003 was genuinely difficult. Geolocation wasn't yet a common part of smallholder agriculture.
"Virtual water"
Their globally weighted average came out to 20,400 liters of virtual water per kilogram of roasted coffee. Virtual water is a concept that goes back to Tony Allan (1993). The idea: every product carries an invisible quantity of water — the water used to produce it. When Switzerland imports wheat instead of growing it domestically, it also "imports" the water that would have been needed to grow it. Virtual water isn't water physically present in the product; it's a purely accounting-based perspective. Hoekstra later built the "Water Footprint" concept on Allan's foundation.
How do you get to 140 liters? How did the authors calculate?
Step 1: How much water does a coffee plant need per year?
The authors use the CROPWAT model, which measures the evapotranspiration of a coffee plant. For Brazil, they arrive at 12,000–13,000 m³ of water per hectare per year — roughly equivalent to a football pitch sitting under about 1.8 meters of water.
Step 2: How much coffee is harvested per hectare?
The authors used FAO average data from 1995–1999 for Brazil, arriving at 1,100 kg of green coffee per hectare. Today we'd assume at least double that. The 13,000 m³ divided by those 1,100 kg gives 11,800 liters of water per kilogram of green coffee.
Step 3: From green coffee to roasted coffee
One kilogram of green coffee yields approximately 0.84 kg of roasted coffee. The 11,800 liters divided by 840 g of roasted coffee gives 14,000 liters per kilogram. The globally weighted figure comes to 20,400 liters per kilogram, because other countries produce significantly less per hectare.
Step 4: The 140 liters
Those 20,400 liters divided by 7 g = 143 liters per cup. The 3-liter difference is rounding.
For a double espresso with 18 g of coffee, you'd get 367 liters.
The core problem with the calculation
The first step is where it matters: the authors ask how much water a planted hectare of coffee evaporates per year and get 13,000 m³. Physically, that's correct. But — and this is what so many people have simply jumped on — that figure makes no distinction between where the water comes from.
The relevant question is: how much additional, scarce water does coffee cultivation actually claim, water that would otherwise be available for something else?
Green, blue, and grey water
The water footprint concept distinguishes three components that have fundamentally different implications for coffee.
Green water:
rainwater stored in the soil and evaporated through the plant. For coffee, this accounts for over 95% of the total — nearly all of those 140 liters. This water falls as rain on the land whether coffee grows there or not.
Blue water:
water drawn from surface water or groundwater — what we actually mean by "water consumption." For coffee, this represents only about 1% of the water footprint (Revolve, 2023, drawing on Mekonnen and Hoekstra 2011). That share has likely grown as more farms have adopted irrigation.
Grey water:
a theoretical construct. It refers to the volume of water that would be needed to dilute fertilizer-related pollution down to acceptable thresholds. This is where you have to look very carefully. Studies from 2011 (Mekonnen and Hoekstra) and 2021 (Leal-Echeverri) differ by a factor of five on this figure, based purely on methodological assumptions.
The underlying problem is that all three water types get added together into one number. It's worse than comparing apples and oranges, because it also adds a theoretical construct into the mix.
The criticism of the concept
Scientific criticism came from multiple directions. Water footprints have not been scientifically validated (Wichelns 2010, 2011), and water scarcity is not a global phenomenon. For coffee specifically, this really matters: coffee is grown where rain falls. Climate change is shifting precipitation patterns, but coffee was never planted in dry regions — it simply wouldn't have grown there.
Since the water footprint was originally developed for irrigated dry regions, not for rain-fed agriculture in humid tropics where most coffee grows, the methodology produces deeply misleading results when applied to coffee (Batchelor 2022).
Following the same method, regional differences become enormous: the authors arrive at 49,000 liters per kilogram in Ghana but only 6,000 in Vietnam. The reason is yield per hectare: the more a hectare produces, the more the water figure gets divided down. Higher yield, lower water use per kilo — that's the logic.
Today's more widely used CO₂ calculations per hectare work the same way, bringing in yield per hectare. More yield, lower emissions — always per unit, per kilogram. We measured this ourselves at Apas.
Processing methods and agroforestry
Wet-processed coffees require water during processing: cherries run through channels, then get depulped and either sprayed with water (ecopulper) or washed in tanks. The water used is fresh water, kept as clean as possible to avoid contamination.
Dry-processed coffees only need fresh water if the cherries are floated in a water channel before drying — the ones that float are removed, as they're less dense and may have a defect.
It's equally unsurprising that coffee grown in full sun uses more water, as it's often artificially irrigated — compared to coffee grown in partial shade or in an agroforestry system where shade trees surround the coffee plants.
Realistic water consumption
In 2003, the researchers had difficulty obtaining precise farm data, so they relied on databases — which produced assumptions rather than certainties. 140 liters per cup sounds catchy and plays well in the media.
But:
how much water does it actually take to produce coffee for one cup? I asked our friends at Apas (Brazil), Cima Coffee (Honduras), Chacra d'Dago (Peru), and Mount Sunzu (Zambia). The short answer: it's not 140 liters.
Mount Sunzu – Zambia: 22 liters
Wet-processed
Luca Costa of Mount Sunzu wrote to me:
«I've looked at the calculations we used to size our irrigation system. The most important point right up front: these are our numbers, for our farm and our processing — a site-specific example, not a general figure for coffee as a whole.»
He's right. That's exactly the point — making specific calculations, analyzing them, and adjusting where needed.
Cultivation/irrigation: 568.5 liters per kg cherry. Processing: 0.9 liters per kg cherry. Green coffee: 3,410 liters per kg. With a roasting loss of 20%, Luca arrives at 32 liters per cup (7 g) — 77% less than the study claims. He's calculating conservatively, erring toward higher figures. He estimates actual on-farm water use is about 30% lower, which would bring the figure to 22 liters per cup (−84% vs. 140 liters).
Yair Keidar, Cima Café, Honduras: < 1 liter
Wet-processed
«Indeed, in Honduras very little if any coffee farms use irrigation, so this is not really a big factor in water usage.» Irrigation is barely used in Honduras. For processing, however, the large majority of Honduran producers use wet processing — and efficiency there depends heavily on the equipment available.
Notably, IHCAFE — the Honduran coffee institute — has picked up the 140-liter figure:
There is a «limit» of 140 liters of water per quintal (bag) that the IHCAFE supposedly enforces, and a producer risks a fine if they exceed this amount of water.
There it is again. Those 140 liters serve as the vague and methodologically questionable basis for a law that could fine producers for using more than 140 liters to process 69 kg of green coffee.
At Finca San José in Santa Bárbara for example, total farm yield is about 80–120 bags (69 kg bags) and they are usually using around 130 liters per bag of washed coffee.
That works out to 1.9 liters per kilogram of green coffee — and for 7 g of roasted coffee, 17 ml. For comparison: the espresso machine itself uses around 60–80 ml of brew water for a double shot. The process water for the entire wet processing stage is less than a quarter of what you pour directly into the machine.
Chacra d'Dago, Peru: < 1 liter
Wet-processed
Fresh water at Chacra d'Dago is only used to process the cherries during wet processing. Water in the flotation tanks, where cherries are sorted, gets cleaned and recirculated.
Based on our records and operational practices, water use can be estimated in a range of approximately 90,000 to 152,000 liters, which is about 0.8 liters of water per kg of cherry under an efficient water management system.
0.8 liters of water per kg of cherry translates to about 35 ml per 7 g of roasted coffee — 99% less than 140 liters. The modern closed-loop flotation system is highly efficient and recycles the water, which would bring the figure down even further.
Apas, Brazil: 0 liters
Dry-processed
At Apas, the coffee plants are not artificially irrigated and the cherries are dry-processed as naturals. That means zero liters of fresh water consumed. What none of these calculations include is the small amount of fresh water used in liquid foliar fertilizer applied to the leaves — but those volumes are minimal and wouldn't change the numbers.
For wet-processed coffees from the region, Mauricio from Apas notes that about 30 liters per 60 kg of green coffee are used for washing, and another 50 liters for depulping. Combined, that's 11.6 ml per cup — again 99.9% less than 140 liters. Or 4.4 ml per cup.
What the real water problems are
The 140-liter figure is no longer relevant, but it refuses to go away. The conversation needs to shift to the water problems that actually affect local ecosystems.
Irrigation in water-scarce regions
It matters closely where the water comes from when coffee is irrigated artificially. According to Brazilian agricultural research agency EMBRAPA, 14% more coffee plantations in Brazil were under artificial irrigation between 2022 and 2024. Wherever high yields are being pushed — Vietnam is another example — water sourcing needs scrutiny.
Wastewater from wet processing
Coffee wastewater contains tannins, phenols, and alkaloids that inhibit biological breakdown. Oxygen gets depleted, creating anaerobic conditions. Untreated wastewater carries a high organic load that is lethal to aquatic organisms and undrinkable. I've visited farms that discharged the agua miel — the wastewater from wet processing — directly into their fields without any treatment. I've also visited farms with state-of-the-art water treatment systems that had the problem under control.
Climate change and shifting precipitation patterns
Coffee's biggest water problem is still ahead of us: Climate Central documents an average of 47 additional days per year above 30°C across the 25 most important coffee-growing countries in 2026. Shade, water retention, and artificial irrigation will keep growing in importance.
