Industries / Energy, water and waste / Environmental services / Stormwater management
Planted and porous surfaces
Soaking up rain where it falls, with planted hollows, green roofs and paving that lets water through, so less of it runs off into drains.
7
Milestones achieved, from 1972 to 2015.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1972
Porous asphalt is tested
Researchers at the Franklin Institute in Philadelphia studied, for the US Environmental Protection Agency (EPA), whether paving with open pores could let rain soak into the ground instead of running off into sewers. Their best laboratory mix was asphalt concrete with 5.5% asphalt by weight, graded so that water passed through at 76 inches (about 1.9 m) an hour, and it held up in tests of stability, durability and freezing and thawing.
Credit
Edmund Thelen, Wilford Grover, Arnold Hoiberg and Thomas Haigh, Franklin Institute Research Laboratories; US EPA
1990
Rain gardens replace ponds
In 1990 the environment department of Prince George's County, Maryland, began trying other ways to handle runoff and developed bioretention, or "rain gardens": shallow planted hollows that collect rain from roofs and streets and let it soak through soil. The county published a bioretention design manual in 1993 and a Low Impact Development (LID) manual in 1997, which spreads many small planted features across a site instead of piping all the water to one pond. By the county's own figures, the developer of one subdivision, Somerset, saved $4,500 per lot, $900,000 in all, by leaving out curbs, ponds and drainage structures.
Credit
Larry Coffman and the Prince George's County Department of Environmental Resources
2009
Toronto requires green roofs
Toronto City Council adopted a bylaw in May 2009, in force from 31 January 2010, that required green roofs (roofs planted over a waterproof layer) on large new buildings, from 2,000 m² of floor area; the city says it was the first city in North America with a bylaw both requiring green roofs and governing how they are built. By the city's count the bylaw led to more than 1,200 green roofs over 15 years. From 3 November 2025 Ontario's government removed the city's power to require them, and a spokesperson for the province's housing minister said voluntary green roofs give builders "the flexibility of choice".
LessonA city's requirement can be withdrawn by a higher level of government, and Toronto's lasted 16 years.
Credit
City of Toronto
2011
Philadelphia plans with green tools
In 2011 the Philadelphia Water Department began Green City, Clean Waters, a 25-year plan to cut the rain entering the old combined sewers that serve about 60% of the city, mostly with rain gardens, tree trenches, porous paving and other planted "green tools" rather than only bigger pipes and tanks. The department says it has built more than 2,800 green tools at nearly 800 sites and passed its 10-year pollution reduction goal.
Credit
Philadelphia Water Department, under a consent order with the Pennsylvania Department of Environmental Protection
2012
Porous parking clogs at edges
In 2009 the US EPA built a test parking lot with three rows of porous surface at its Edison Environmental Center in New Jersey. Infiltration tests through August 2012 found clogging along the uphill edges, where runoff and grit from the ordinary paved driving lanes flowed onto the porous rows. At an 80-space school lot at Fort Riley, Kansas, EPA's buried sensors showed the clogged band creeping 1 mm across the surface for every 6 mm of rain.
LessonPorous paving clogs first where dirty water from hard surfaces runs onto it, so what drains onto it matters as much as the paving itself.
Credit
Thomas O'Connor, Michael Borst and Mahdi Razzaghmanesh, US EPA
2015
Soil filters save salmon
In tests by Washington State University and NOAA Fisheries, untreated runoff from a Seattle highway killed juvenile coho salmon and the insects they eat, and passing the same runoff through bioretention soil removed the lethal effect. Once the killer was traced to a tire chemical (see Water quality), a 2023 field test on a mature bioretention cell, extended with a model, found such cells cut the load of 6PPD-quinone reaching streams by more than about 90% in typical storms, those up to the size expected once in two years.
LessonA planted filter can be judged by whether fish downstream survive, not only by how much water it holds back.
Credit
Jenifer McIntyre, John Stark and colleagues (Washington State University); Nathaniel Scholz (NOAA Fisheries); Timothy Rodgers, Rachel Scholes and colleagues (2023 study)
2015
China starts sponge cities
In 2015 China's State Council launched "sponge city" pilots, which a member of the Chinese Academy of Engineering calls China's version of low impact development; its October 2015 guidelines say that by 2030 about 80% of urban areas should absorb and reuse at least 70% of rainwater. Thirty cities joined the pilots. A 2023 review found that some sponge cities still flood, and a 2024 model study of a district in Wuhan found the measures worked for short storms of the size expected about once a year but did less as storms grew.
LessonSoaking up everyday rain does not stop a cloudburst from flooding a city.
Credit
State Council of China and the pilot cities
Watch for
Philadelphia meets its 15-year target
The Philadelphia Water Department says that at year 15 of Green City, Clean Waters it has passed the "greened acre" standard in its consent order with Pennsylvania's environment department: more than 3,860 greened acres from over 1,100 projects, against 3,812 required. It also says the projects built or under way will exceed its overflow-reduction target.
Achieved whenthe Pennsylvania Department of Environmental Protection accepts the year-15 report, or an outside review confirms the acres and the overflow reduction they deliver.
Claimed by
Philadelphia Water Department, July 2026
Open
01
A sponge city scorecard
A public table of measured results for China's 30 pilot sponge cities: for each pilot area, the share of yearly rain it actually held back, from monitoring rather than from models, set against the 70% target, with the source page for every number.
The briefCredit · unclaimed02
Toronto after the repeal
A running count of new Toronto buildings that still get green roofs now that they are voluntary, set against the 15 years when they were required, with a check from aerial photos that the roofs listed in permits were built.
The briefCredit · unclaimed
Open 01 · the brief
A sponge city scorecard
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public table of measured results for China's 30 pilot sponge cities: for each pilot area, the share of yearly rain it actually held back, from monitoring rather than from models, set against the 70% target, with the source page for every number.
Why it is still open
A 2023 review found that some pilots still flood, and much of the published evidence is model studies of single districts, such as the 2024 Wuhan study. Some pilots do publish monitoring: Zhenjiang's designers report ten years of records from 40 automatic stations across its 29 km² pilot area, with 76% of yearly runoff held back against a 75% target (Novatech conference, 2026). Searches in English and Chinese found such single-city reports, but no open, compiled table of measured results for all the pilots.
Built with what exists
China's national standard for assessing sponge city results (2018); open papers found through Crossref and Europe PMC; city and provincial evaluation reports; a web agent to gather the documents (see the Note "Deleting 80% to start over"); an AI model that reads Chinese to pull out figures, with every figure checked against its page by a person; a spreadsheet.
First step
List the 30 pilot cities and their pilot areas, and collect each city's published evaluation reports.
Done looks like
All 30 pilots listed, each with either a measured figure, its rainfall years and how it was monitored, or a note that none is published; 20 entries checked by a second reader of Chinese, with the disagreements published.
Open 02 · the brief
Toronto after the repeal
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A running count of new Toronto buildings that still get green roofs now that they are voluntary, set against the 15 years when they were required, with a check from aerial photos that the roofs listed in permits were built.
Why it is still open
Ontario removed the requirement from 3 November 2025. The city's open green roof permit data, downloaded on 18 November 2026, lists 55 permit applications from 2024, 21 from 2025 and 4 from 2026, but recent years may be incomplete, and searches found no study that separates a real drop from slow recording or measures what the repeal has done.
Built with what exists
The City of Toronto's open "Building Permits - Green Roofs" data, refreshed weekly, and its full building permit data; aerial images whose license allows analysis; a small vision model fine-tuned to tell a planted roof from a bare one (see the Note "One GPU to fine-tune"); Python.
First step
Download the green roof permit data and the full permit list, and plot green roof permits as a share of large new building permits by month.
Done looks like
Monthly counts from 2010 to the present, as a share of permits for large new buildings, updated every quarter for two years after the repeal; and a photo check of 100 buildings finished since 2023, with the model's agreement with a person's labels published.