Capturing and treating drinking water

Dams, wells, reagents, filters, sludge, testing and public health protection.

Making a changing water safe

Three sources, three ways of capturing

Capturing water from a river requires screens to keep out debris and fish, intakes designed to work at both high and low flow, and pumping stations able to adapt to a flow that varies with the seasons and rainfall. Capturing from an underground aquifer requires drilling wells and, above all, protecting that resource against overexploitation: extracting water faster than the aquifer naturally recharges can deplete it almost irreversibly on a human timescale, or cause saltwater intrusion into coastal aquifers. Capturing from a reservoir requires actively managing the sediment that accumulates behind the dam and the algal blooms that can appear in stagnant, warm water. In all three cases, the quality of the source water depends on decisions made long before capture — what is allowed to be discharged upstream, how the watershed's land is managed —, and no treatment plant, however advanced, can indefinitely compensate for a systematically polluted watershed.

Inside the plant: a sequence of physical and chemical barriers

The standard drinking-water treatment process follows a sequence of stages, each designed to remove a different type of impurity. First, the water's chemistry is adjusted and coagulants are added that make the finest suspended particles clump into larger, heavier flocs. Those flocs settle by gravity in large sedimentation tanks, and the resulting water then passes through filters — sand, anthracite or other materials — that retain the remaining particles. The final barrier is disinfection, usually with chlorine or other disinfectants, which eliminates pathogenic microorganisms that survived the previous stages. When the source water has specific contaminants — pesticides, certain organic compounds, tastes or odors —, additional stages are added, such as activated carbon, finer filtration membranes, or other specialized treatments designed for that particular contaminant.

None of this is free, and nothing simply disappears

Each of these stages consumes chemical reagents, electricity for pumps and aeration systems, compressed air in some processes, water to periodically wash saturated filters, and constant laboratory analysis to verify the treated water meets required health standards. And it is essential to understand that the contaminants removed do not disappear in the process: they end up concentrated in sedimentation sludge, in the reject brines of some treatments, or in saturated filters that need replacing — all that contaminant mass keeps existing after treatment, just in a much more concentrated and manageable state than when it was diluted in the raw water.

Disinfection has to protect all the way to the last meter of pipe

It is not enough to disinfect water as it leaves the plant: that protection has to be maintained until the water actually reaches each home's tap, often through kilometers of distribution network. Old pipes, made of degraded materials or with internal biofilms built up over decades, can re-contaminate water even after it has already been properly treated at the plant — which is why network operators continuously monitor residual disinfectant levels, pressure and microbiological quality at various points in the network, not just at the treatment plant's outlet. At the same time, excess disinfectant generates its own unwanted chemical byproducts, so the technical goal is always the minimum level of disinfection needed to guarantee safety to the farthest point of consumption, no more and no less.

Prioritizing demonstrated risk, not the week's headline

Managing drinking-water safety requires systematically prioritizing risks with solid scientific evidence and known serious consequences — biological pathogens, chemical contaminants with demonstrated toxicity at the concentrations present — over whatever substance receives the most media attention at a given moment, which may or may not represent a real risk at the concentrations typically found. This does not mean ignoring emerging contaminants or stopping research into them: it means allocating limited monitoring and treatment resources according to the available evidence on real risk, not according to which headline generates the most alarm that particular week.

The cheapest barrier is upstream

In favor of protecting the watershed over expanding the plant: preventing discharges, protecting aquifer recharge zones, controlling agricultural nutrient use, and conserving watershed soil directly reduces how much chemical and energy-intensive treatment is needed afterward.

To watch at the plant itself: excess disinfectant generates unwanted chemical byproducts; insufficient disinfectant lets pathogens through. The technical goal is always the minimum needed to guarantee safety to the farthest point of consumption, no more and no less.

Always compare: source water quality, energy consumed, reagents used, waste generated (sludge, brines, saturated filters) and the whole system's demonstrated reliability — "natural water" or "spring water" does not automatically mean safe water without treatment.

The mistake of chasing the week's headline: always prioritize risks with solid scientific evidence and known serious consequences, not whichever substance gets the most media attention at a given moment.

The numbers, with their source

More than 40% of treated water never gets billed

Globally, so-called "non-revenue water" — water that is captured and treated but lost before reaching a paying user, through physical leaks or metering errors — exceeds 40% of drinking water produced, equivalent to about 45 million cubic meters a day. The World Bank recommends keeping this figure below 25%; countries with highly modernized networks, such as Japan or the United Kingdom, fall below 15%, while some developing regions exceed 50%. Every cubic meter of that non-revenue water represents reagents, pumping energy and treatment work invested without anyone ever actually using that water — this section's leaks and open-tap leaf develops this cost in more detail.

Sources: World Bank — The Challenge of Reducing Non-Revenue Water; Aquatech — Essential Guide: Leakage & Non-Revenue Water.

Why this figure matters so much for reading the rest of this section

Every liter of non-revenue water is not just an economic loss for the network operator: it is also a real energy loss, because that liter already went through capture, chemical treatment and pumping before being lost. A system with 50% non-revenue water is, in practice, treating and pumping twice as much water as actually reaches users — a fact that completely changes how any figure for "treatment energy consumption" should be read without also knowing the specific network's loss rate.

What this table is missing

This global figure mixes two very different causes: real physical leaks in aging pipes, and "commercial" losses from illegal connections or poorly calibrated meters — both count as "non-revenue water," but require completely different engineering solutions.

Sources