part of EQUORA Institute
HUEN
EQUORA Institute · Active research thread · Water

Scroll to the end and try every interaction → earn a Resonator Pass. 5 Passes = free attendance at an invitation-only Interference evening.

Mental Fast Food™ · Data Visualisation

Global freshwater supply
to 2030

One of humanity's most critical challenges: different in form by region, but globally interconnected. Drought, glacial melt, shifting monsoons, reservoir evaporation, desalination — interactive forecast to 2030.

Source
UN · UNU-INWEH · World Bank · FAO (Food and Agriculture Organization) · IPCC (Intergovernmental Panel on Climate Change)
Collaboration
EQUORA Institute × GoHalve
Updated
July 2026

Interference topic · The invisible water loss

What if water scarcity isn't a quantity problem — but a question of who decides where the rain falls?

Click a region on the map

Interactive map — click a region

Challenge type: Drought Glaciers Monsoon Desalination Evaporation
Global forecast · 2030
50%
of humanity will live in water-stressed areas
Of all human water use
85%
goes to agriculture — only 15% remains for industry and drinking water
Hidden loss
~7%
of freshwater lost to evaporation from reservoirs and dams
Regions covered
10
areas facing distinct types of water crisis

Context

Why
now?

The water crisis is not a future threat — it is already underway. According to the IPCC 6th Assessment Report, climate change impacts water systems most severely: glacial melt, increasingly extreme precipitation patterns, and massive reservoir evaporation together cause dramatic supply declines.

The GoHalve × Equora collaboration aims to make these interconnected processes visually comprehensible — with region-specific context, not global averages.

85% of human water use goes to agriculture, while ~7% of global human water consumption evaporates annually from the world's reservoirs and dams — typically not counted in "consumption" figures.

"But doesn't evaporated water just rain back down as fresh water?" Not where it's needed. Yes, water cycles globally — but evaporation from a reservoir in a dry region is a local loss that doesn't return there. Much of it falls over oceans or already-wet regions, and climate change is shifting where rain lands — often away from the places running short. Add the water locked in glaciers (now melting and draining to the sea) and the water polluted beyond use. The freshwater actually available, where and when people need it, keeps shrinking. By 2030, current trends suggest half of humanity will live in areas where water demand exceeds natural replenishment.

Source: IPCC AR6 · FAO AQUASTAT · World Resources Institute · Nature Climate Change 2023

2026 · UN update

The water ledger
goes bankrupt

In January 2026, ahead of the UN Water Conference, the UN University's water institute (UNU-INWEH) published its flagship report Global Water Bankruptcy, and it moves the story above one step further: the world has entered an era of global water bankruptcy. The report speaks the language of finance, because it treats water systems as an account. Rivers, soil moisture and snowmelt provide the annual, renewable income; aquifers, glaciers and wetlands hold the long-term savings. Many societies have by now overspent both — they exhausted the annual income, and along the way began drawing down the savings as well.

Bankruptcy here is a precise, two-part condition, fixed in the peer-reviewed definition paper behind the report. Insolvency means withdrawals and pollution persistently exceed renewable inflows and the safe limits of depletion. Irreversibility means parts of the natural capital — aquifers, wetlands, soils, glaciers — can be restored on human timescales only at prohibitive cost, if at all. Together they describe a system that keeps functioning, only on a different, lower baseline.

The balance sheet, in numbers

50%
of large lakes losing water since the early 1990s
A quarter of humanity depends directly on those lakes.
70%
of major aquifers in long-term decline
Half of domestic water worldwide is now drawn from groundwater; 40%+ of irrigation comes from aquifers being steadily drained.
410 M ha
of natural wetlands erased in five decades
Roughly the area of the entire European Union.
30%+
of global glacier mass lost since 1970
Entire low- and mid-latitude ranges are expected to lose functional glaciers within decades.
100 M ha
of cropland damaged by salinization alone
Dozens of major rivers now fail to reach the sea for parts of the year.
~4 bn
people live with severe water scarcity at least one month a year
Nearly half the global population.

The baseline has moved

For planning, the report's central conclusion concerns baselines. Many basins and aquifers have settled into a persistent failure state: they can still be operated sustainably, yet they can no longer return to their historical normal. This confirms, at UN level, the conclusion Gergő reached when he measured Hungary's water account in Balatons: planning has to start from new baselines, because the historical averages on which water rights, irrigation schemes and urban growth were built are the imprint of a climate that is measurably gone. Sizing for the »Old Normal« amounts to borrowing against an income that no longer exists.

Try it: the water account

»Old Normal« is the amount of water that renewed each year under the earlier, stable climate — the level irrigation, water rights and city planning were all built around. It no longer exists: less water arrives now, yet plans still assume the old figure. The model below keeps the "account" of a single aquifer from 2026 to 2100. Three decisions shape it: how much we withdraw relative to the »Old Normal«, how much climate change reduces the annual income, and in which year we align withdrawals with the actual yield. The lesson draws itself as you play: an early turn preserves the baseline, a late turn merely stabilises a lower one, and a turn after bankruptcy amounts to damage management.

Water account simulator · 2026–2100

savings irreversibility threshold

It has to be solved — the directions that follow

The report is emphatically a call to action, because the financial logic of bankruptcy also marks out the way through it: whoever is bankrupt first draws up an honest balance sheet, then aligns spending with income, and finally protects the remaining capital. Translated to water, that means four directions.

1

Operate within the hydrological envelope. Water allocations need to be tied to the actual, annually re-measured renewable income, and renegotiated periodically — fixed, decades-old water rights are promissory notes of the old climate.

2

Demand is the fastest lever. Agriculture takes the bulk of withdrawals, so efficient irrigation, crop switching and consumption choices deliver the largest savings together — the footprint of individual products can be explored, with adjustable weights, in the beverage footprint map and in which water-saving move matters.

3

Measure what you manage. Keeping an account without bookkeeping is impossible; dense, continuous monitoring of groundwater levels, soil moisture and small watercourses is therefore the first infrastructure of bankruptcy management — the same "we measure, we don't estimate" direction we are building at Equora.

4

Soil and wetlands are the rebuildable savings. Glaciers are lost on human timescales; humus-rich soil and restored wetlands, however, can store water again — the everyday side of soil water retention is shown in the composting piece, the regional stakes in the drying of the Carpathian Basin.

A question for Iterators™ at a future Interference™

If the water account is an account, who holds signing authority — what vote would we give the next two generations over today's withdrawals?

Source: Madani, K. (2026): Global Water Bankruptcy — Living Beyond Our Hydrological Means in the Post-Crisis Era. UNU-INWEH, doi: 10.53328/INR26KAM001 · Madani, K. (2026): Water Bankruptcy — The Formal Definition. Water Resources Management 40(78), doi: 10.1007/s11269-025-04484-0 · Litkai Gergely: A hazai vízkassza Balatonokban. LinkedIn, 6 July 2026

What you can do

Not the
tap.

Here's the trap: turning off the tap while brushing feels responsible, but it's a rounding error. The water you really use, you never see — it's hidden inside what you buy. Agriculture takes 85% of human water use, and most of your footprint is "virtual water" embedded in your food, your clothes, your devices.

First, the water you do see — straight from the tap, per person, per day
~125
litres / day
Germany
~330+
litres / day
United States (indoor)
<20
litres / day
Much of the Global South

Hold that number — roughly 125 litres a day for a typical European. Now watch how many days of it are hiding inside a single thing you buy.

Gone in one sitting

196
litres
A single egg
= ~1.5 days of tap water
~250 eggs a year per person — that's where the litres stack up, not in one breakfast.
130
litres
One cup of coffee
= ~1 day of tap water
~125 l is the bean, grown in Ethiopia or Brazil — not your tap. And if you drink one a day: ~47,000 l a year.
34
litres
One cup of tea — ~4× less than coffee, but still 136 cupfuls
Same morning ritual, 4× smaller footprint. This is one of the few swaps where the choice actually matters.
~30
litres
A 0.5l Coca-Cola — mostly the sugar, not the bottling
The company said 2 l. The independent measurement said ~30 l. Who measured it, and what did they leave out? →The number behind the number
~1 700
litres
One 100g chocolate bar — 11 bathtubs
= ~14 days of tap water
~1,600 l is the cacao — mostly grown in West Africa, not a water-stressed region. The number is real; the crisis it implies is less straightforward than beef.
15 400
litres / kg
Beef — one steak ~2,300 l
one steak = ~18 days · one kg = ~123 days
This one is real across all three dimensions: large volume, frequent consumption, and often grown in water-stressed regions (US Great Plains, Brazil's Cerrado). The single highest-impact food choice.
16 000
litres / kg
Almonds — a driver of California's drought
one kg = ~128 days of tap water
If you're buying almonds in Europe, they're almost certainly from California — one of the most water-stressed agricultural regions on earth. This is a direct link between your shopping basket and the drought.
5 000
litres / kg
Cheese — five times milk
one kg = ~40 days of tap water
The concentration effect: it takes ~10 l of milk to make 1 kg of cheese. Average European consumption is ~20 kg/year — that's ~100,000 l annually, just from cheese.

Lasts you years

2 700
litres
One cotton T-shirt
= ~22 days of tap water
Worn 3 years (150×): ~18 l per wear. Worn 3 months and discarded (15×): ~180 l per wear. The purchase moment decides, not the washing.
10 000
litres
One pair of jeans
= ~80 days of tap water
Worn 5 years (500×): ~20 l per wear. Fast-fashion cycle, replaced every season (30×): ~333 l per wear. Same number, completely different impact.
12 760
litres
A smartphone — each chip rinsed 30+ times
= ~102 days of tap water
Used 5 years: ~7 l/day of ownership. Replaced every 2 years: ~17 l/day. The upgrade cycle is the water decision — not how you charge it.
2–3
litres
A plastic bottle — more than it holds, and gone in minutes
Less than an egg — but daily, every day, discarded. At one bottle a day: ~900 l/year, ~750 discarded objects. The volume is small; the accumulation is not.

But not all litres are equal. A T-shirt, jeans, a phone cost a lot up front — yet they last years, so the water spreads thin across every wear. An egg, a coffee, a chocolate bar are gone in one sitting, and tomorrow you buy another. The plastic bottle is the worst of both: a durable-goods footprint for something thrown away in minutes. The question isn't just "how much water" — it's "how much, how often, and for how long."

And recycling itself?

"Just recycle it" sounds like the clean answer — but recycling drinks water too, and water is precisely the yardstick on which the gain disappears. In energy and greenhouse gases recycled PET is clearly better than virgin; the US industry life-cycle comparison, however, found water consumption essentially the same for the two routes. The heavy item is washing, and the measured values scatter widely with plant and technology: figures from ~3.5 l/kg to ~27 l/kg appear in the literature for mechanical recycling. Without a closed-loop wash circuit, what is gained in material and energy can melt away on the water side. Which opens a genuinely uncomfortable question: what is worth recycling, in water terms — and could we end up with mountains of recycled material while running dry?

Where is this researched?

And the power itself

Even electricity has a water footprint — mostly for cooling. And the surprise: the "cleanest"-looking source is the thirstiest. A hydropower reservoir loses ~7–50 litres of water per kWh to evaporation, while solar sips ~0.1 l/kWh (just panel-washing) and wind almost none. Coal and nuclear sit around 2 l/kWh. So "renewable" and "low-water" aren't the same question — which is exactly the kind of measured detail worth getting right. Note: these figures are only the operating footprint of finished infrastructure — the water to build the dam, panel or turbine is extra. Moving the power down the grid, by contrast, needs essentially no water: transmission lines don't cool or consume.

Where is this researched?

1

See it first. The numbers above already shift the question from "did I leave the tap running?" to "what did I eat and buy this week?" That's where the litres actually are. Awareness isn't the goal — it's the starting line.

2

Measure, don't estimate. A vague "I should use less" changes nothing. A measured footprint — yours, not a national average — shows you the two or three choices that carry most of the weight. Usually: less beef, fewer fast-fashion purchases, more attention to what's grown in water-stressed regions.

3

Then halve what you can. Not everything — the few things that matter, measured against your own baseline. Halving one heavy category beats a dozen symbolic gestures.

This is where the open question begins — and it's bigger than one person's habits. Most footprint numbers are estimated, not measured. Measuring your real impact, the water behind power and recycling, then halving what counts — these are open research directions we're building at Equora.

EQUORA Institute → GoHalve →

Key data

Open research direction

The water behind
your electricity

How much freshwater each kilowatt-hour really costs — split by source (solar, wind, hydro, nuclear, gas), and separating the water to run the infrastructure from the water to build it. Today these numbers are mostly estimated and averaged. The aim: measure them honestly, per source and per region.

This is a direction we're developing — it doesn't have its own page yet. "We measure, we don't estimate."

EQUORA Institute →
Open research direction

What's worth recycling,
in water terms?

Recycling saves material — but the washing stage drinks water, and for some materials the full life-cycle water cost can rival making it new. The open question: which materials genuinely pay off in water terms, so we don't end up with mountains of recycled stuff while running dry?

This is a direction we're developing — it doesn't have its own page yet. "We measure, we don't estimate."

EQUORA Institute →
The number behind the number

Who measures,
and how?

For years the company mainly communicated a tidy ratio — about 2 litres of water per litre of drink — which counts only the bottling. But the independent Water Footprint Assessment put a half-litre bottle at ~28–35 litres, once the sugar in the field is included. The factory is a rounding error; the crop is the footprint.

It went further: during the study, the company pushed to adopt a "net green" accounting trick that would have erased ~43% of the figure. The researcher, Arjen Hoekstra — who created the water-footprint method — refused. The lesson isn't "Coke is uniquely bad." It's that a number means little until you know who measured it, and which parts they left out.

TakeawayA number is worth little until you know who measured it and what they counted in. Before accepting or passing on a figure, it pays to find the boundary of its method: on the same question, counting only the bottling and counting the whole production chain differ by an order of magnitude.

Read the documented investigation →
Same research thread The Carpathian Basin is drying → Which water-saving move matters → The footprint of drinks → Track your own footprint — GoHalve →
Research provenance
This page comes out of research at the EQUORA Institute and captures one state of that work rather than a settled institutional position. That state rests on the findings available at the time of publication; later findings appear here only where the page has been updated, which the date shows. AI takes part throughout the research process as a thinking partner; responsibility for interpretation and publication remains human.
Published: 15 June 2026 · Updated: 19 July 2026 · Last content update: 17 August 2026
Papp László · EQUORA InstituteHow We Research →
Scientific background
Freshwater scarcity, virtual water, and the 2030 horizon — the research base

The research is free — and will stay free. If it matters to you, there's a way to say so.

Support the work →