Drainage & the Water-Sensitive Site

To design water-sensitive sites you first have to see how ordinary development manufactures flooding, because the mechanism is so routine that it is almost invisible. The root of it is the replacement of pervious ground - soil, vegetation, open earth that absorbs and delays rain - with impervious surfaces - roofs, roads, car parks, paved compounds and compacted plots that absorb almost nothing. Two things change when you do this, and both make flooding worse.

First, the volume of runoff rises sharply. On natural ground a large share of rainfall infiltrates into the soil or is taken up by plants, and only a fraction runs off. Pave that ground and almost all the rain becomes surface runoff, so the same storm now produces far more water rushing across the site and into the drains. Second, the runoff arrives faster and more peaked. Soil and vegetation slow water down and release it gradually over hours; smooth hard surfaces and piped drains speed it up, so the runoff from a developed catchment reaches the drains and watercourses in a sudden, concentrated peak rather than a gentle, spread-out flow. A higher peak arriving sooner is exactly the recipe for overwhelming a drainage system.

Stack this across a whole neighbourhood and you get the familiar urban flood: every plot sheds its rain quickly onto a drainage network that was often undersized to begin with, is frequently silted, blocked or built over, and sometimes discharges into watercourses and lakes that have themselves been encroached and narrowed. The water has nowhere to go and nothing to slow it, so it ponds in the streets and enters the buildings. Indian cities have seen this vividly - rapid development covering the soft ground and the natural drainage lines that once absorbed and carried the monsoon, turning manageable rain into repeated flooding. The liberating insight, and the subject of the rest of this lesson, is that because development causes this, design can reverse it: if each site sheds less water, more slowly, and soaks up more, the cumulative effect is to relieve the flooding that development created. The designer owns the decision to manage water on site; the quantified runoff and drainage sizing belong to the drainage engineer and the authority.

Sustainable drainage systems - SuDS - are the design response to the impervious-surface problem, and their philosophy is a direct reversal of the old approach. Conventional drainage tried to get rain off a site as fast as possible through pipes and gutters, pushing the problem downstream; SuDS tries to mimic natural drainage by managing rain close to where it falls - slowing it down, storing it temporarily, letting it soak into the ground, and releasing only a reduced, delayed flow to the drains. The aim is to bring a developed site's runoff closer to what the natural ground would have produced, so development adds little to downstream flooding.

SuDS works through a management train - a sequence of measures that treat water in stages from source to destination, so no single element has to cope with everything. Source control comes first: deal with rain where it lands - green roofs that hold and slow rainfall, permeable paving that lets it soak through, water butts and rainwater harvesting that capture roof water for reuse. Next, conveyance and storage that slow and hold water rather than rushing it: swales (shallow planted channels) that carry water slowly and let it infiltrate, rain gardens and bioretention areas that collect and soak runoff, and detention or retention basins and ponds that store a storm's peak and release it gradually. Finally, infiltration where the ground allows: soakaways and permeable areas that return water to the soil rather than to a pipe. Each stage slows, stores and cleans, and what leaves the site is a fraction of what bare paving would have shed.

SuDS carries real co-benefits that make it attractive beyond flood control: it recharges groundwater (precious where water tables are falling), improves water quality by letting soil and plants filter pollutants, cools and greens the site, and creates habitat and amenity. This is the same instinct behind the 'sponge city' idea gaining ground internationally and in Indian urban thinking - designing the city to absorb and hold water like a sponge rather than shedding it like a roof. For the designer the message is practical: on every site, ask how rain can be slowed, stored and soaked rather than rushed away, and build a management train suited to the plot. The suitability of infiltration (which depends on soil and water table), and the sizing of every SuDS element for the design storm, are matters for a drainage engineer and the local authority - treat the principle as yours and the numbers as theirs.

Two site moves do an outsized share of the work, and both are squarely in the designer's hands: choosing permeable surfaces over impervious ones, and grading the ground so water goes where you want it.

Permeable surfaces attack the impervious-surface problem at its root by letting rain soak through the surface into the ground or a storage layer below, instead of shedding it. Permeable paving, gravel, porous blocks, reinforced grass and simply keeping more of a plot as planted, open earth all let the ground keep drinking. The design discipline is to minimise the impervious footprint - question every square metre of hard paving, keep driveways and compounds permeable where you can, break up large paved areas with planting, and avoid the reflexive Indian habit of tiling or concreting the entire plot, which turns a garden into a rain-shedding roof. Less hard surface, and more of it permeable, directly reduces the runoff the site produces.

Grading - the shaping of ground levels - decides where water goes once it is on the site, and it is astonishing how often it is got wrong. The first rule is simple and non-negotiable: grade the ground to fall away from the building on all sides, so rain and runoff drain away from the walls and foundations rather than ponding against them or running back toward the building. A building set in a local low spot, or ringed by paving that slopes toward it, will be wet at its base and may flood through its lowest openings even in ordinary rain. The second is to lead water deliberately to where you want it - to the swales, rain gardens, soakaways and storage of the management train, or to the public drain - along routes that are planned, not accidental. The third is to protect the building's threshold: keep the finished floor above the surrounding ground and the design flood level, and make sure surface water cannot pond at the door.

Together, permeable surfaces and good grading mean a site that produces less runoff and sends what remains safely away from the building and into measures that slow and soak it. These are low-cost, high-value decisions made in the earliest site layout - exactly the designer's territory. How much a site must attenuate, whether the soil will infiltrate, and the levels and falls for the design storm are confirmed by the drainage engineer and the authority; the instinct to let the ground drink and to fall water away from the building is yours to hold from the first sketch.

The final shift is to stop treating rain as only a threat to be shed and start treating it as a resource to be managed - which, happily, aligns almost perfectly with reducing flood risk. The same measures that slow and store runoff can capture it for use, and in much of India, where water is scarce for months and then arrives all at once, this double benefit is compelling.

Rainwater harvesting is the clearest example: capturing roof runoff in tanks for reuse, or directing it to recharge pits and soakaways that return it to the groundwater, simultaneously reduces the peak runoff leaving the site and banks the water for the dry season or the aquifer. Many Indian cities now mandate rainwater harvesting on new plots for exactly this reason, and done well it is both a flood measure and a water-security measure. Detention storage that holds a storm's peak and releases it slowly, green roofs that hold and evaporate rain, and rain gardens that soak and filter it all play the same dual role: less flooding downstream, more water kept in the landscape. The designer's move is to integrate these into the scheme from the start - sized and sited as part of the building and landscape, not bolted on - so managing rain becomes part of how the project works.

Underlying all of it is a principle of neighbourliness that flood design makes unavoidable: water is a shared, connected system, and a site's drainage decisions land on other people. A plot that paves itself over and sheds its rain fast is exporting its flood risk to the street and the houses downhill; a plot that fills a natural drain or blocks an overland flow path to gain a little space can flood its neighbours. Conversely, a site that slows, soaks and stores its own rain is quietly protecting everyone downstream. Good drainage design is therefore partly an ethical act - not making your water someone else's problem - and partly a civic one, because when many sites do it, the cumulative effect genuinely relieves urban flooding. This is where building-scale flood design meets the city, and where the designer's choices, plot by plot, add up to whether a neighbourhood floods or not. The binding requirements - mandatory harvesting, permitted discharge rates, the treatment of natural drains and flow paths - come from the local authority and the drainage engineer; the disposition to manage rain on site and be a good hydrological neighbour is the designer's to carry into every project.