Flood resilience is usually measured in flooded roads avoided, drainage capacity added and property damage prevented. A new peer-reviewed study suggests that cities may need to add another line to that ledger: public health.
Published in Nature Cities on 8 September 2026, the study analysed approximately 4.2 million bacillary dysentery cases across China from 2005 to 2020. It found that flood exposure increased disease risk, while cities implementing China’s sponge city initiative showed lower flood-related bacillary dysentery risk. The largest reported reduction was 12.2% by the third year after implementation.[1]
The finding arrives during another reminder of how quickly urban drainage can become a health and infrastructure issue. In mid-September 2026, heavy rain caused fatal flooding and transport disruption in eastern Spain, submerged more than 2,000 homes in northern Vietnam, and contributed to deadly flash flooding on the Navajo Nation in New Mexico.[3][4][5] At the same time, NOAA reported that El Niño was strengthening, with a greater than 90% chance of a very strong event during Northern Hemisphere fall and winter 2026–27 and a 75% chance that the October–December three-month Relative Oceanic Niño Index would reach at least +2.5°C.[2]
That does not mean these individual floods were caused by El Niño. It does mean that planners, engineers and developers are operating in a period when extreme rainfall preparedness deserves renewed attention — including the health consequences that follow when stormwater and sanitation systems fail together.
Key takeaways:
- A 2026 Nature Cities study linked sponge city development with a 12.2% reduction in flood-related bacillary dysentery risk by year three.[1]
- Flood exposure, especially severe flooding, increased bacillary dysentery risk over the post-flood lag period.[1]
- School-age children and older adults were among the more vulnerable groups identified in the study.[1]
- The protective association was stronger in cities with weaker baseline infrastructure, including fewer public toilets and lower water-pipeline density.[1]
- The study strengthens the case for evaluating green stormwater infrastructure not only as flood control, but also as infrastructure capable of delivering potential public-health co-benefits.
Why Sponge City Public Health Benefits Matter
Floodwater in a city is rarely just rainwater. When drainage, sewerage and sanitation systems are overwhelmed at the same time, runoff can mix with sewage, contaminated surface water and waste from flooded sanitation facilities.
The World Health Organization has highlighted this pathway for years. In a 2023 Bulletin of the World Health Organization article on sponge cities, WHO experts noted that around 1.7 billion urban residents lacked safely managed sanitation, with many relying on pit toilets or septic tanks rather than sewer connections; these systems can be among the first overwhelmed during flooding.[6]
Bacillary dysentery is particularly relevant because it is commonly associated with Shigella bacteria and spreads through the faecal–oral route. Symptoms can include diarrhoea, sometimes bloody, stomach cramps and fever.[7] A systematic review of 141 epidemiological studies also found broad evidence that diarrhoeal disease tends to increase following heavy rainfall and flooding events.[8]
This makes stormwater management a public-health issue as well as an engineering one. When urban flooding increases contact between people, contaminated water and failed sanitation systems, reducing the frequency, depth and duration of that exposure can plausibly reduce health risk.
What the 2026 Nature Cities Study Found
The new study, led by researchers from Lanzhou University and Tsinghua University, used two complementary approaches. First, the researchers analysed approximately 4.2 million bacillary dysentery cases across China between 2005 and 2020 to estimate the acute relationship between flood exposure and disease risk. Second, they used a quasi-experimental design to compare changes in flood-related disease risk between cities with and without sponge city development.[1]
| Study element | Finding |
|---|---|
| Study period | 2005–2020 |
| National disease dataset | Approximately 4.2 million bacillary dysentery cases |
| Flood effect | Flood exposure increased bacillary dysentery risk, with stronger effects from severe floods |
| Vulnerable groups | School-age children were more vulnerable to overall flooding; older adults were especially vulnerable to severe flooding |
| Sponge city effect | Lower flood-related bacillary dysentery risk in cities with sponge city development |
| Largest reported reduction | 12.2% by the third year after implementation |
| Equity pattern | Stronger protective effects where baseline infrastructure was weaker |
The most important wording here is risk. The study did not report that sponge cities simply produced 12.2% fewer total dysentery cases in every setting. Rather, the quasi-experimental analysis estimated a 12.2% reduction in flood-related bacillary dysentery risk by the third year of the initiative.[1]
That distinction matters for both scientific accuracy and project evaluation. The result is strong evidence that sponge city development can deliver measurable health co-benefits, but it should not be interpreted as a universal disease-reduction guarantee for every city, climate or infrastructure configuration.
How Flooding Creates a Disease Pathway
The health pathway can be understood as a chain:
- Extreme rainfall exceeds local drainage capacity.
- Stormwater interacts with sanitation systems, including combined sewers, septic tanks, pit toilets or contaminated surface runoff.
- Contaminated water reaches streets, homes, public spaces or water sources.
- People are exposed through direct contact, contaminated food or water, or poor hygiene conditions after flooding.
- Enteric disease risk rises in the days and weeks following the event.
A sponge city strategy attempts to interrupt this chain earlier — before runoff becomes concentrated enough to overwhelm downstream infrastructure.
How Green Stormwater Infrastructure Can Interrupt Exposure
China’s sponge city approach combines multiple forms of source control, detention, infiltration, filtration, reuse and drainage. In practice, this can include green roofs, rain gardens, bioswales, permeable pavements, wetlands, detention areas and underground storage.
No single component can prevent every flood. The value comes from a distributed management train in which different measures reduce runoff at different points in the catchment:
- Green roofs intercept rainfall at the building level and delay the rate at which water reaches roof drains.
- Bioretention and rain gardens temporarily store runoff while supporting infiltration and filtration.
- Permeable surfaces reduce rapid surface runoff where site conditions allow infiltration.
- Detention and underground storage provide additional capacity when source-control measures are exceeded.
- Conventional drainage remains essential for safe conveyance, overflow routes and events beyond the design standard.
This is why the most useful comparison is not “green versus grey.” In many projects, the more realistic model is blue–green–grey integration: distributed source control supported by engineered drainage and safe overflow capacity.
Why the Year-Three Result Should Not Be Overinterpreted
The Nature Cities study found its largest reduction in flood-related bacillary dysentery risk in the third year after sponge city implementation.[1] That timing is important, but the study does not prove that vegetation establishment, substrate biology or maintenance learning caused the three-year pattern.
From an engineering perspective, several factors could reasonably contribute to improving performance over time: more projects may be completed across the drainage network; vegetation and soil-based systems can mature; maintenance teams gain operational experience; and weak points such as clogged inlets or poorly connected overflow routes can be identified and corrected. These are plausible mechanisms, not findings established by the epidemiological study itself.
For project owners, the practical lesson is that sponge city performance should be evaluated over multiple years rather than judged from a single storm season. Useful long-term indicators may include:
- percentage of planned source-control area completed;
- runoff volume and peak-flow performance;
- vegetation survival and substrate condition;
- inspection frequency and clogging rates;
- overflow and nuisance-flooding records;
- maintenance response time; and
- where public-health agencies have suitable data, post-flood enteric-disease trends.
Adding Public Health to the Stormwater ROI Ledger
Stormwater investment is normally justified through avoided flood damage, reduced sewer overflow, regulatory compliance and increased drainage capacity. Public-health co-benefits are harder to monetise, so they are often left outside the business case.
The 2026 study suggests that omission may undervalue distributed green infrastructure. If a stormwater programme reduces the frequency with which polluted floodwater reaches people, then avoided illness, healthcare use and lost productivity are legitimate benefits to investigate — even if local values must be calculated separately.
There is already evidence that green infrastructure can change infrastructure economics. A 2025 analysis of Kansas City’s Smart Sewer programme reported a 50-year lifecycle cost of approximately US$1.0 per gallon of annual overflow captured for a green-infrastructure alternative, compared with about US$8.6 per gallon for the relief-sewer option it replaced in that specific project.[9] This should not be generalised as a universal cost ratio, but it demonstrates why multi-benefit alternatives deserve project-specific comparison rather than being treated as decorative additions.
A more complete climate-adaptation business case can therefore examine at least four benefit streams:
| Benefit stream | What to measure |
|---|---|
| Flood reduction | Peak flow, runoff volume, flood depth, flooded properties and disruption hours |
| Water quality | Overflow frequency, suspended solids, pollutant loads and receiving-water impacts |
| Public health | Post-flood enteric-disease incidence, healthcare use and exposure in vulnerable populations |
| Co-benefits | Heat mitigation, biodiversity, amenity, urban greening and property-level resilience |
Where Modular Green Roofs Fit in a Sponge City Strategy
For dense developments, rooftops are one of the few large urban surfaces where rainfall can be intercepted before it reaches street-level drainage. Modular green roofs can therefore serve as one source-control layer within a wider sponge city or green stormwater infrastructure strategy.
For shallow extensive applications, the HOENSOEY HT-508 green roof module uses a 500 × 500 × 80 mm profile with adjustable water-retention and drainage components. It is intended for shallow planting applications where a low-profile modular system is preferred, subject to project-specific structural, waterproofing and drainage review.[10]
For deeper planting, the HOENSOEY HT-5020 semi-intensive green roof tray uses a 500 × 500 × 200 mm structural tray profile with interlocking connections and multi-directional drainage channels. The 200 mm dimension is the tray profile, not a universal growing-medium depth; substrate depth, saturated weight, vegetation, irrigation and structural loading still need to be designed for each project.[11]
Where additional subsurface detention is required, HOENSOEY Cells provide geocellular components with a published void ratio of approximately 95%. Published compression figures are component references and should be checked against the relevant test report, assembly and project design before being used as engineering design values.[12]
The key point is system integration. Green roof trays, roof outlets, emergency overflows, waterproofing, planting media, underground storage and downstream drainage must be coordinated as one hydraulic system. A high-performing component cannot compensate for a blocked outlet, inadequate overflow path or poorly maintained interface.
Specification Lessons From the Health Evidence
The public-health finding does not create a new product specification on its own, but it changes what long-term performance means. If part of the value of sponge city infrastructure is reducing exposure to contaminated floodwater, then systems must remain functional well after installation.
Four specification priorities follow:
- Design for maintainability. Provide accessible outlets, inspection points, removable components and clear maintenance procedures.
- Verify hydraulic interfaces. Source-control measures need safe downstream conveyance and emergency overflow routes.
- Use project-specific structural values. Tray dimensions, empty weights or component compression tests are not substitutes for a complete saturated-load and structural assessment.
- Track performance beyond handover. Inspection and maintenance records should be treated as part of infrastructure performance, not as optional landscaping documentation.
These principles matter because green infrastructure only delivers its intended flood, water-quality and potential health benefits while it remains hydraulically functional.
Conclusion: The Sponge City Health Dividend Is Becoming Measurable
The most important result from the 2026 Nature Cities study is not simply the headline 12.2%. It is the emergence of measurable public-health evidence around a type of infrastructure that has traditionally been justified through drainage, ecology and urban design.
The study found that flooding increased bacillary dysentery risk and that cities with sponge city development experienced a lower flood-related risk, with the largest reported reduction — 12.2% — appearing by the third year after implementation.[1] The stronger protective effect in places with weaker baseline infrastructure also suggests that the value of nature-based adaptation may be especially important where existing sanitation and drainage defences are limited.
For engineers and municipal planners, this does not replace hydraulic design standards. For developers and building owners, it does not mean that a green roof alone can prevent flood-related disease. What it does provide is a stronger reason to evaluate green roofs, bioretention, permeable surfaces, storage and conventional drainage as parts of one resilience system — and to include public health among the outcomes worth measuring.
References
- Zhang, L., Zha, Q., Zhao, H. et al. “Reduced flood-related bacillary dysentery risk with sponge city development in China.” Nature Cities, published 8 September 2026. https://www.nature.com/articles/s44284-026-00514-z
- NOAA Climate Prediction Center. “ENSO Diagnostic Discussion,” 10 September 2026. NOAA CPC ENSO discussion.
- Xinhua, republished by The Star. “Heavy rains in eastern Spain leave one dead, disrupt transport,” 18 September 2026. Source.
- Associated Press. “Heavy rain and floods in northern Vietnam submerge homes and disrupt economy,” 18 September 2026. Source.
- Associated Press. “3 dead after being swept away by floodwaters on the Navajo Nation,” 17 September 2026. Source.
- Thompson, F. “Sponge cities for healthier populations.” Bulletin of the World Health Organization, 2023;101(2):86–87. Source.
- UK Health Security Agency. “What is Shigella? How do you catch it and how can it be avoided?” 26 March 2026. Source.
- Levy, K., Woster, A.P., Goldstein, R.S. & Carlton, E.J. “Untangling the impacts of climate change on waterborne diseases.” Environmental Science & Technology, 2016;50(10):4905–4922. Source.
- Tomić, S. “Adaptive management approach – Kansas City success story.” Cambridge Prisms: Water, 2025. Source.
- HOENSOEY. “Green Roof Module HT-508.” Product page.
- HOENSOEY. “Semi-Intensive Green Roof Tray System HT-5020.” Product page.
- HOENSOEY. “HOENSOEY Cells: Shallow Drainage Cells and Internal Components.” Product page.