Explainer: Sponge cities make room for rain to soak in, collect temporarily or move through planned floodable spaces. They can reduce pressure on drainage networks, but a planted park does not make a neighbourhood flood-proof. The useful question is where water goes during a storm—and what happens when the available space fills.
For Southeast Asian readers, Singapore’s Bishan–Ang Mo Kio Park offers a concrete example of this approach. It also shows why a successful project combines landscape design with drainage engineering. This article explains established approaches; it is not a report of a new flood or a newly announced project.
What does a sponge city actually do?
The phrase describes an approach to urban water management rather than one piece of equipment. Rain gardens, planted drainage channels, permeable surfaces and storage areas can manage rain closer to where it falls. The US Environmental Protection Agency describes green infrastructure as using vegetation, soils and other measures to store, infiltrate or return stormwater to the atmosphere, reducing flows into drains and waterways. Read the EPA overview.
Three different jobs are easily confused:
- Infiltration: water enters the ground where soil and groundwater conditions allow.
- Temporary storage: water occupies a basin or other designed space before draining or being used.
- Conveyance: water travels along a route designed to carry it. A floodable river corridor can perform this job without absorbing all the water into its soil.
These distinctions matter when judging a photograph. A green riverbank is not evidence that rainfall is disappearing underground. A temporarily wet park may be using its intended flood space rather than demonstrating that a project has failed.
Singapore’s example: a park that gives the river room
At Bishan–Ang Mo Kio Park, PUB and NParks replaced a concrete canal with a naturalised river under the ABC Waters programme. PUB explains that the channel connects to the city’s drainage network and follows a floodplain concept. Water stays in a narrower stream during dry conditions; during storms, adjacent park space becomes part of the route carrying rainwater downstream. Plants, rocks and engineering techniques also help stabilise the banks. See PUB’s project description.
The lesson is to ask what space has been deliberately reserved for water. This is an example of a compatible water-sensitive design approach, not a claim that Singapore adopted another country’s named sponge-city programme. It also does not establish a flood-reduction percentage for another city.
A simple calculation: how much room does rain need?
Consider an illustrative site covering 10,000 square metres, or one hectare. A rainfall depth of 50 millimetres is 0.05 metres. Multiplying the two gives:
10,000 m² × 0.05 m = 500 m³ of rainwater.
Suppose a proposed storage area has 200 m³ of usable capacity. That equals 40% of the rain falling on this example site. It does not establish a 40% reduction in flooding.
This is an original dimensional example, not a measurement or forecast for Singapore. It ignores water arriving from outside the site and is not a drainage design. Actual runoff depends on surfaces, infiltration and existing storage. The timing of rain and discharge also matters: 50 millimetres in a short burst can create a different peak flow from the same depth spread across many hours.
When someone advertises a project’s storage volume, ask for the contributing catchment area and the storm used in the calculation. Without them, a large number in cubic metres says little about protection.
Where the approach reaches its limits
The ground may not absorb water quickly
Clay or compacted urban soils can limit infiltration. Some sites therefore need underdrains or designs that store and convey water rather than relying on it soaking away. Groundwater conditions and pollution risks also affect suitability. These are site-specific engineering choices, not a reason to assume every green space will work equally well. EPA’s design guidance explains these constraints and the need to plan maintenance.
Storage is finite, and the overflow still needs a route
Rain gardens and ponds can slow or hold runoff, while protected floodplain space can complement conventional infrastructure. EPA distinguishes local flooding that overwhelms drainage from flooding caused by rivers exceeding channel capacity. Both require understanding the wider water system. Its flood-mitigation guidance emphasises modelling and floodplain management.
For a proposed project, the practical follow-up is: after the designed capacity is exceeded, does water have a planned route, or does it reach occupied buildings? A landscaped surface cannot answer that question by itself.
Maintenance is part of performance
Planting is only the beginning. Maintenance plans must identify tasks, staff, access and resources. A project that cannot be maintained may not deliver its intended function. Budgeting only for construction leaves out an important part of the system’s working life.
Five questions to ask about a local proposal
The following is an editorial checklist drawn from the distinctions above, not a certification scheme:
- Which flood problem is it addressing? Ask whether the target is local stormwater, river overflow or another source of water.
- What area drains into it? Compare the project footprint with the much larger area that may send it runoff.
- What storm and starting conditions were modelled? Look for rainfall duration and assumptions about available storage.
- Where does excess water go? Ask for the overflow route and how people and property are protected.
- Who maintains it and checks results? A named operator, funded inspections and measurable outcomes are more useful than a rendering of a green park.
Compare a proposal with its stated purpose. Temporary water in a designed floodplain and damaging water in a home are different outcomes. Neither a dry-weather photograph nor a single storm photograph is enough to quantify a project’s benefit.
How this fits a broader climate-ready city
Flood planning and heat planning both involve decisions about streets, buildings and public space. They should be assessed together without assuming that one intervention solves every problem. Our urban heat island explainer covers the separate questions involved in cooling cities.
Water management also has a different meaning on farms. For that distinction, read our guide to alternate wetting and drying in rice production. Agricultural irrigation practices should not be treated as an urban drainage blueprint.
The takeaway: assess a sponge-city proposal as a connected water system. Look for suitable soils or alternative drainage, enough usable space, a planned overflow route and maintenance that continues after opening day. The value is in demonstrated performance under stated conditions—not the label alone.