A standard 250-gram cotton T-shirt has an estimated total water footprint of about 2,720 litres in the Chapagain and Hoekstra calculation. That total combines roughly 1,110 litres of green water (rainfall), 1,230 litres of blue water (surface or groundwater) and 380 litres of dilution water associated with pollution; those categories do not have identical local impacts.
That figure is the accepted estimate from the World Wildlife Fund. It includes the water used to irrigate the cotton crop, the water consumed in dyeing and chemical treatment of the fibres, the water lost in fabric processing at the mill, and the water used in every downstream stage before the finished garment reached a retail shelf. If a single person drank two litres of water a day for two and a half years — as the World Health Organization recommends — they would consume the same volume the T-shirt did before it was ever worn.
The number is difficult to hold in the imagination. It is one of those statistics that becomes immediately abstract on contact with the mind. Two thousand seven hundred litres. A T-shirt. Two and a half years of somebody’s drinking water. The units don’t quite reconcile, and the impulse is to move past them to something more comfortable.
Which is part of why the fashion industry has been able to keep operating on this scale for so long.
Where the water actually comes from
The T-shirt in question is unlikely to have been produced anywhere near you.
The cotton in a typical Western consumer’s clothing is grown in specific regions of the world that have become, over the last several decades, the industry’s chosen production geography. The three largest cotton-producing countries are China, India, and the United States, followed by Pakistan, Brazil, Uzbekistan, and a scattering of smaller producers. The fabric processing that turns raw cotton into finished cloth happens most intensively in Bangladesh, Vietnam, Cambodia, Indonesia, and again China and India.
These are not incidental locations. They were chosen for specific reasons — cheap labour, permissive environmental regulation, existing infrastructure, and, quietly, water. Cotton is one of the most water-thirsty crops on Earth, and cotton processing is one of the most water-intensive industrial activities in modern manufacturing. A production geography for cotton has to be a geography with water.
The problem is that the specific water-rich areas that once made cotton production viable are, in many cases, no longer water-rich. Some of them have been depleted by the industry itself.
The Aral Sea in Central Asia — once the fourth largest lake in the world — has essentially disappeared over the past sixty years, primarily because the two rivers that used to feed it were diverted for cotton irrigation in Uzbekistan and Kazakhstan. What was once open water the size of Ireland is now a salt desert with a few residual pools of water in what used to be its centre. The disappearance of the Aral is one of the largest single acts of ecological destruction ever attributable to a single industry.
In Bangladesh, one of the largest garment-manufacturing hubs in the world, textile factories consume approximately 1.5 trillion litres of water annually. This is groundwater that is not being replaced at anything close to the rate it is being extracted. Aquifers are dropping. Wells are running dry. The people who live in the same towns as the mills are experiencing water shortages driven, in significant part, by clothing being made for consumers on the other side of the world.
What 79 billion cubic metres actually looks like
A UK parliamentary report, citing the 2017 Pulse of the Fashion Industry report, put the global fashion industry’s annual freshwater consumption at an estimated 79 billion cubic metres. The figure is an industry-wide estimate, not a direct measure of cotton irrigation alone.
Seventy-nine billion cubic metres is enough water to fill Loch Ness, Scotland’s largest lake by volume, approximately ten times over. It is enough water to meet the total annual freshwater needs of the entire United Kingdom, twice, with a substantial amount left over. It is, by any reasonable estimate, enough water to provide every one of the 2.2 billion people currently lacking safe drinking water access with the World Health Organization’s daily minimum consumption for several decades.
That last comparison is the one that lands hardest. The industry is consuming, every year, a volume of water sufficient to fully resolve the world’s largest ongoing humanitarian water crisis for a period measured in decades. And the industry is doing so, in significant proportion, in the same countries where that humanitarian water crisis is at its worst.
The 2.2 billion people currently without safe drinking water access are not distributed randomly around the world. They are concentrated in regions that overlap significantly with the fashion industry’s production geography. The same countries whose citizens cannot reliably access clean water are the countries whose cotton is being grown, whose fabric is being dyed, whose finished garments are being shipped to consumers who will wear them for an average of a few dozen wears before disposing of them.
Why the T-shirt matters
The T-shirt is, of course, just one garment. The claim is not that any individual purchase is a moral disaster. The claim is that the industry’s aggregate operation is drawing enormous volumes of fresh water from regions that cannot spare it, and doing so at a scale that is difficult to fully appreciate until the specific geography is understood.
Fast fashion — the industry model that dominates the sector as of 2026 — has intensified this problem substantially over the last two decades. Global clothing production has approximately doubled since 2000. The average garment is worn a fraction as many times as it was in the 1990s before being discarded. The industry’s water consumption has grown in step, with no reduction in per-garment water intensity to offset the increase in volume.
The 2,700-litre figure per T-shirt is not decreasing. Some newer production techniques — waterless dyeing, closed-loop fabric processing, regenerative cotton practices — do exist, and are being adopted in small numbers by some brands. But the industry-wide water footprint is still climbing, because volume is climbing faster than efficiency.
What consumers actually have
The question of what to do with this information is genuinely hard, because individual consumer decisions do not, in aggregate, solve industry-wide problems. Buying less does help. Buying second-hand helps considerably. Choosing brands that publish full supply-chain water footprints helps a small amount. Choosing garments in fabrics with lower water intensity — linen, hemp, some synthetics with the caveat that they create different environmental problems — helps at the level of the individual purchase.
None of this is sufficient to change the industry’s aggregate behaviour. That change, if it comes, will come through regulation, disclosure requirements, and pressure on brands to reveal what most of them currently do not — where specifically their water is being taken from, and what has happened to that place since they started taking it.
Where cotton is grown matters. Rainfall, irrigation withdrawals and pollution-dilution estimates carry different consequences, and a global product footprint cannot show that every T-shirt depleted a particular river or aquifer.