We keep buying water from the tanker trucks

Water and climate series · Part 2 of 3

1. The mountain is the refuge · 2. Water is the condition · 3. Mineral, compute and water

Part one established that the mountain protects us from heat and that the water sustaining it has been declining for forty years. This part is about what can be done, and about why Cochabamba has a full reservoir and neighbourhoods buying water by the truckload at the same time.

Read part 1 if you landed here first.

Cochabamba: water above, thirst below

Cochabamba has the advantage of not depending on glaciers. Misicuni fills with rain and runoff from the Tunari catchment. The reservoir holds 175 million cubic metres and today sits at ninety percent. In an average year it delivers 3,100 litres per second: two thousand for human consumption and eleven hundred for irrigating more than three thousand hectares. Enough for over a million people.

The metropolitan area already passes 1.4 million inhabitants. And of those two thousand litres per second, around forty percent is actually distributed.

That is not scarcity. It is an unfinished public work. The difference matters because one is solved by rain and the other by budget and decision.

On 28 June 2026 the JICA interconnection was delivered, linking two systems and carrying Misicuni water to more than 24,000 families in the south. The missing piece, according to the IDB representative at the ceremony, was 2,300 metres of pipe. The original project dated back to 2007, with Japanese investment of 12.4 million dollars.

Nineteen years between the financing and the tap. The physical distance was 2,300 metres.

Phase 3, capturing the Viscachas and Putucuni catchments, would double supply and take the system past two million people. It must be operating by 2033 and is still seeking financing. The company acknowledges the main challenge is social, because it means intervening in territory held by communities living in those catchments.

Water is not lost. It is badly distributed

There is a second hole, older than climate change and cheaper to plug.

In 2016 the Association of Emeritus Engineers of Cochabamba presented the system balance at the forum marking twenty years since the Water War. One thousand litres per second available, technical losses of forty percent, meaning four hundred litres per second, 12.6 million cubic metres a year escaping through the network. In money of the time, some 5.6 million dollars evaporating every year through broken pipes.

The same document lists the recurring pattern: a pumping station on the Titiri river built for a million dollars and abandoned; deep wells costing six million that produced no water; a canal from neighbouring catchments losing fifty percent of fifty percent to infiltration. Works built, paid for, and never maintained.

And where the state did not arrive, people drilled. The balance counts around 1,250 wells in the valley, extracting some forty million cubic metres a year. Five times Escalerani, a third of Misicuni's first stage, pulled from underground with no shared plan. The word the engineers use for that figure is alert.

It is the same aquifer water that amunas and qochas are trying to recharge. Every new well competes with infiltration from above.

Ten years of life

This is not a metaphor. It is a measurement.

Ciudatos Bolivia and Lab TecnoSocial published research in 2023 on urbanisation and inequality in the country, written and visualised by Valeria Peredo. They take Avenida Heroínas in Cochabamba and show that for many people it works as a border between the north and the south of the city.

To the north, in districts 1, 2, 3, 4, 10, 11 and 12, most people have drinking water and sewerage. To the south, in districts 5, 6, 7, 8, 9, 14 and 15, a large share of the population still gets water from tanker trucks or from precarious wells, and thousands of families still have no sewerage. The south also holds the wastewater treatment plant and the municipal dump.

The figure that closes the argument comes from a 2013 study the publication draws on.

The distance between those two numbers is the same distance separating the tank with 945 litres per second of supply and zero demand from the neighbourhood buying water by the truckload.

And the pattern is not only Cochabamba's. The research shows every Bolivian city repeating the same shape: a privileged centre and a marginalised periphery. Dual cities, built by the people of those peripheries themselves since the 1980s, with no planning and no services.

When climate pushes more people toward the valleys and the altiplano, they will arrive in that periphery. Not in the north. Climate inequality will stack on top of an urban inequality that already exists and is already measured in years of life.

The technology we already have

The obvious answer to a disappearing glacier is a reservoir to replace it. A reservoir captures the peak and cannot hold it all if the whole year arrives in three storms. It also costs hundreds of millions, takes fifteen years and displaces people.

There is another layer of infrastructure, cheaper and faster, designed a thousand years ago for exactly a regime of rain concentrated into a few months. That is the regime we are heading into.

Amunas

A canal captures rainwater above 4,400 metres and carries it to zones of fractured rock, where it infiltrates instead of running off. The water travels slowly inside the mountain and emerges between two and eight months later through springs, fifteen hundred metres lower, right in the dry season.

Qochas

Natural depressions in the terrain that hold rain and runoff, with capacity extended by an earth dike. Those left unlined do double duty: they store visible water for livestock and irrigation, and at the same time recharge the aquifer below, extending spring flow through the dry season.

Bofedales

High altitude wetlands. They work as a sponge: absorbing in the rainy season and releasing slowly the rest of the year. Herders extend them by diverting water from streams and springs through channels, which means much of the Andean bofedal is hydraulic infrastructure built by people, not untouched natural landscape.

The three, compared

System	Where the water goes	How long until it returns	What it needs
Amuna	Fractured rock, deep aquifer	Two to eight months	Annual communal work, canal cleaning
Qocha	Surface and shallow aquifer	Weeks to months	Earth dike, basin maintenance
Bofedal	Peat and saturated soil	All year, continuous seep	No overgrazing, no draining
Reservoir	Visible impoundment	Immediate, subject to evaporation	Hundreds of millions, fifteen years, displaced people

The first three are the counterpart of the lost ice. They do not replace a glacier's full volume, and they do what the glacier did and the reservoir does not: regulate timing. Store when there is surplus, deliver when there is none.

The part that is rusting

Here the circle closes, and it is the uncomfortable part for a city that tells itself it won the Water War.

Research by Anna Ramoneda Martí, published in the Anuario del Conflicto Social nineteen years after the conflict, reached a hard conclusion. The idea of water managed as a commons surged during the Water War, faded over time and has now disappeared completely. The reason the study identifies: social movements and civil society lost influence against formal decision-making bodies.

The city won the battle against privatisation and then lost, slowly and without a fight, the collective capacity to decide about its own water.

That matters technically, not only politically. The community water systems of the southern zone, grouped in Asica Sur, want to keep managing in a decentralised way and at the same time need to receive Misicuni water. How that water stored in the cordillera gets distributed, and with how much decentralisation, remains unresolved.

An amuna needs communal labour every year. A bofedal needs grazing agreements. A shared well needs someone enforcing the limit. Without that social layer, the infrastructure exists and does not work.

Anthropologist Fánel Guevara put it best: without an organised community, amunas are not possible. Andean hydraulic infrastructure is social infrastructure. It collapses when the organisation collapses.

What needs doing now

Order matters, because some things take a decade and others can start this year.

1. Finish the networks that already have water allocated. There is flow bought and paid for that never reaches houses. The cheapest and fastest item on this list.one to three years
2. Plug the leak. Forty percent losses means nearly half the city's hydraulic effort goes into the ground. Repairing old network costs a fraction of a new reservoir and delivers water the same year.one to four years
3. Regulate the wells. 1,250 boreholes pulling forty million cubic metres a year, with no shared registry or limit. Cadastre, metering and an agreed rule.start this year
4. Measure. Everything. Spring flows, aquifer levels, bofedal cover, real consumption per neighbourhood. Without a public time series there is no way to know whether an intervention worked.start this year
5. Rehabilitate amunas, qochas and bofedales at catchment scale. Cost per cubic metre far below a reservoir, two to five year timeline, and the benefit reaches upstream communities and downstream cities at once.two to five years
6. Write the agreements before the works. Phase 3 and any future transfer touch community territory. Conflict arrives when the agreement is negotiated after the machines are already in.before 2028
7. Protect the upper catchment from mining and land speculation. A contaminated aquifer or a subdivided headwater cancels everything above.permanent
8. Phase 3 operating in 2033. That is the date the company itself sets. Financing closed before 2028 or it does not arrive.2033

What is at stake

By mid-century, around 1.5 billion people in lowlands will depend critically on water coming down from mountains. The Andes feed the Amazon. What happens above four thousand metres does not stay up there.

And a migration has already begun. If a third of the planet falls outside the habitable climate niche, people move toward where the body can function. The Andes are one of those places on this continent. We will receive population, with or without a plan.

A mountain city with water sorted receives people and grows. A mountain city without water sorted becomes a place where the climate permits living and the infrastructure does not. That is the difference between the two possible Cochabambas, and it is decided in the next ten years.

The mountain already did its part. It gave us 2,560 metres of thermal advantage for free. All that is missing is storing the water and getting it there, and the climate does not decide that. We decide it, now, or the lack of water decides it for us, later.

Part 3 Mineral, compute and water. Two global industries arrive to bill water in the same territory, in the same decade.