Your city does not flood because it rained more. It floods because the water has nowhere left to go.

Every monsoon, an Indian city goes under water and the explanation offered is that it was an unprecedented downpour. Sometimes it genuinely was. Far more often, the same city absorbed a comparable storm three decades ago without much incident, and what has changed is not the rain but the ground it lands on. Urban flooding is mostly a land-surface problem wearing the costume of a weather event, and the hydrology behind it is straightforward enough that the failures become obvious once you can see them.

Rain that used to soak in now runs off

Drop rain on undisturbed soil with vegetation and most of it never becomes a flood. Some is intercepted by leaves, some infiltrates and recharges groundwater, some is held in surface depressions, and only a fraction runs off. Engineers describe this with a runoff coefficient — the proportion of rainfall that becomes surface flow. For open ground with good soil it might be one in ten; for a concreted, roofed, asphalted urban block it approaches nine in ten.

Two things change together as a city hardens, and the second is the one people miss.

The first is volume. Nearly all the rain that falls now has to travel across the surface, so the drainage network is asked to carry several times what it once did.

The second is speed. Water on a rough, vegetated slope moves slowly; water on a paved street moves fast, and roof-to-gutter-to-drain is faster still. Hydrologists call the time it takes for runoff from the far edge of a catchment to reach a given point the time of concentration, and paving collapses it. The consequence is that runoff from the whole area arrives at the same place at the same time instead of trickling in over hours. The peak flow does not merely grow — it grows and arrives sooner and sharper. A drain sized for a gentle rise fails against a spike.

This is why flooding worsens in a locality where nobody built anything dramatic. Enough courtyards concreted over, enough plots roofed, enough verges paved, and the catchment quietly changes character.

The storm the drain was designed for

Every stormwater network is built to a design storm — a rainfall intensity of a stated severity and duration, usually expressed by return period. Indian urban drains have commonly been designed for something in the range of a two- to five-year storm, which sounds negligent until you consider the cost of sizing an entire city's pipes for the worst rainfall in fifty years. Some flooding, in any honest design, is accepted.

Two things then go wrong.

The city outgrows the assumption. A network laid for a colony of bungalows with gardens now drains apartment blocks with parking podiums. The design storm was never revised; the runoff coefficient doubled.

The rainfall record stops describing the future. Design storms are drawn from historical intensity–duration–frequency curves, which assume the statistics of rainfall are stable. Over India that assumption is weakening: research on the observational record finds a marked rise in widespread extreme rainfall events over central India in recent decades, even where total seasonal rainfall has not risen. A drain designed correctly against a 1980s rainfall record can be undersized today without a single error in the calculation. Total monsoon rainfall is a poor predictor of flooding; what floods a city is intensity — how much falls in three hours.

The storage a city threw away

Conveyance is only half of drainage. The other half is storage — places water can sit while the peak passes.

Lakes, tanks, marshes and floodplains are storage. They are also flat, well-drained-looking, unbuilt and centrally located, which is exactly what makes them attractive to build on. Across Indian cities the same sequence has repeated: tank beds become bus stands and stadiums, marshes become IT parks, floodplains become housing, and the natural channels between water bodies become roads. Each conversion is locally reasonable and collectively fatal, because the buildings are not merely in the way of the water — they occupy the volume the water used to occupy.

That volume does not disappear. It relocates to whatever is lowest nearby, which is why the flooded street is so often the one built on a former channel, and why residents notice that water "comes from nowhere" in a particular lane every year.

Roads make it worse in an unglamorous way. Each resurfacing raises the carriageway a few centimetres, and after several cycles the road sits above the plinth of the houses along it. The drainage then works exactly as designed, into the living rooms.

Why bigger drains disappoint

The reflexive response to flooding is to build larger drains, line them with concrete and straighten them. This helps at the point of complaint and frequently makes the system worse.

A smooth, straight, concrete channel moves water faster. Faster water reaches the downstream junction sooner, where it meets everything else that has also been sped up, and the peak downstream gets higher. Conveyance-only thinking pushes the flood along rather than reducing it — a well-known result in urban hydrology and one that Indian cities keep rediscovering, neighbourhood by neighbourhood.

Coastal cities face a hard limit that no amount of pipe diameter solves: tidal locking. A drain discharging into the sea cannot discharge against a high tide. When heavy rain coincides with a high tide, the outfalls are effectively shut and the city has to store the water regardless of how good the network is. Pumping stations and tide gates help, and they are storage-and-timing solutions rather than capacity solutions.

Then there is the failure that requires no hydrology at all. Drains silted, encroached, roofed over, or serving as informal sewers and waste dumps do not perform at their design capacity — often at a fraction of it. Plastic waste in particular ends up at the inlet gratings, and a network can be sized perfectly and still fail at a screen the width of a footpath.

A city does not need to convey a storm out. It needs to hold it long enough for the peak to pass. Every square metre of ground that can absorb or store water is drainage capacity that nobody has to maintain.

What actually works

The design shift now widely accepted is from moving water away as fast as possible to slowing, spreading and soaking it — sustainable drainage, low-impact development, or in its best-known form, the sponge city.

In practice this means permeable paving in low-traffic areas, bioswales and planted verges that intercept road runoff, rain gardens, retention and detention basins that fill during a storm and drain slowly afterwards, restored tanks and wetlands, and rooftop rainwater harvesting — which recharges groundwater and, in aggregate, removes a real share of runoff from the network at source. Cities that mandated harvesting, Chennai notably, did it primarily for water supply; the drainage benefit is a side effect worth counting.

The most effective interventions are the least visible. Protecting an existing wetland outperforms any engineered basin built to replace it, and floodplain zoning — restricting what may be built where water is known to go — is the cheapest flood control there is. India has had a model floodplain zoning bill circulating since the 1970s that few states have enacted, and the National Disaster Management Authority issued dedicated urban flooding guidelines in 2010 recommending catchment-based planning rather than ward-based drain building. The knowledge is not the missing piece.

Why it matters for students and researchers

Urban flooding is a genuinely interdisciplinary problem — hydrology, urban design, planning law, materials, governance — and it is unusually accessible to student research. Catchment delineation and runoff modelling can be done with open elevation data and free tools; historical land-cover change is visible in freely available satellite imagery going back decades; and the most valuable datasets of all, records of where water actually stood and how deep, are largely uncollected in Indian cities and can be built from municipal complaints, news reports and residents' own accounts.

The open questions are practical: how should design storms be revised when the rainfall record is no longer stationary; how much retention capacity does a dense Indian neighbourhood need and where can it physically go; how do informal settlements, which carry the largest share of flood harm, fit a planning framework that often does not acknowledge them; and what maintenance regime actually keeps a drain at design capacity in a city with informal waste flows.

That is squarely the territory of the International Journal of Urban Design and Development (ISSN 2584-0568), a peer-reviewed hybrid open-access journal launched in 2023 covering urban nature and city design alongside urban design methods and techniques. For architecture, civil engineering and planning students, urban flooding is a good reminder that the decisive design decision is often not the building but the ground it replaced.

Frequently asked questions

Why do cities flood even when rainfall has not increased?

Because paving replaces soil that used to absorb rain. Runoff volume rises sharply and reaches the drains far faster, so the peak flow is both larger and earlier than the network was designed for, even for the same storm.

What is a design storm?

The rainfall intensity and duration a drainage system is built to handle, usually stated as a return period. Indian urban drains are commonly designed for a two- to five-year storm, which means some flooding in rarer events is accepted by design.

Does building bigger drains solve urban flooding?

Only locally. Larger, smoother channels move water downstream faster, raising the peak where they discharge. Reducing flooding usually requires storage — retention basins, wetlands, permeable surfaces — rather than conveyance alone.

Why does flooding happen where lakes and marshes used to be?

Because those water bodies were the city's storage, and buildings occupy the volume that water used to fill. The water still arrives and moves to the lowest available ground, which is frequently the filled tank bed or the former channel now used as a road.

What is a sponge city?

An approach that slows, spreads and absorbs rainfall rather than channelling it away — using permeable paving, bioswales, rain gardens, retention basins, restored wetlands and rainwater harvesting to hold water near where it falls.

Why do coastal cities flood at high tide?

Because outfalls cannot discharge against a higher sea level. When heavy rain coincides with a high tide, drains back up regardless of their capacity, and the city must store the water until the tide falls or pump it out.