BRENNEN

Industries / Energy, water and waste / Environmental services

Stormwater management

Stormwater management is the work of handling the rain that falls on towns: collecting it in gutters, drains and pipes, holding it in ponds, tanks and planted ground, cleaning out what it picks up from streets and roofs, and warning people when it floods.

c. 600 BCE

Rome began the Cloaca Maxima, a drain that carried the flood-prone stream of the Velabrum valley to the Tiber and dried the marshy ground where the Forum was built; Rome's heritage office says Livy places the start of the work in the 6th century BCE, under Tarquinius Priscus.

sovraintendenzaroma.it

36

Milestones achieved across 6 domains.

9

To watch for: announced, not yet confirmed outside.

15

Open deliverables. Nobody has shipped them.

In one minute

Where it started

Rome built the Cloaca Maxima from about the 6th century BCE to drain the marshy valley under its Forum, and in 33 BCE Agrippa had Rome's drains cleaned and sailed through them. Engineers still size most small drains with the rational method, a one-line formula that Emil Kuichling applied to city sewers in 1889, using rain statistics such as the US Weather Bureau's 1961 atlas of 10- and 100-year storms. From the late 1960s, US towns also made new building sites hold back their runoff in ponds.

Where it stands

US towns have needed permits for the runoff their storm drains carry since 1987, more than 2,200 of them charge a stormwater fee by Western Kentucky University's 2026 count, and the US Environmental Protection Agency's 2022 survey put the country's stormwater needs at $115.3 billion over 20 years. Cities now soak up rain with rain gardens, green roofs and porous paving, from Philadelphia to China's sponge city pilots, and send what the pipes cannot carry along streets and through tunnels, as Copenhagen and Kuala Lumpur do. In England, rain on towns now threatens more properties than rivers and the sea: 4.6 million against 2.4 million, by the Environment Agency's 2024 count.

What is open

Much of what is missing is evidence rather than tools: nobody publishes how flash flood warnings score against street sensors, whether old ponds still hold what their permits required, or which towns still size pipes for the rain of the 1950s. The open jobs below use public records, AI models that read documents and photos, and open runoff models such as EPA's SWMM to fill those gaps, and to test the AI tools now sold to drainage reviewers, flood forecasters and maintenance crews.

Domains

AchievedWatch forOpen

Open in stormwater management

3 of 15 open deliverables

Open · Drains and pipes

A pipe check for towns

A free tool that screens every storm pipe in a town for whether it can carry today's design storm. It reads the town's public map of storm pipes (diameter, slope, material), draws the area draining to each pipe from lidar elevation data, takes rainfall from NOAA Atlas 14, and applies the rational method and Manning's formula for pipe flow to flag pipes that would run over full.

Built with what exists

QGIS; WhiteboxTools to draw drainage areas; USGS 3DEP lidar; NOAA Atlas 14 rainfall; Python with hydro-tools; EPA SWMM to check a sample; an AI coding agent to write the glue code.

Credit · unclaimed

Open · Ponds and holding areas

An exam for drainage reviewers

A public test for AI tools that check the drainage reports developers submit to towns. It holds real reports with the comments town reviewers sent back, covering runoff, pipe sizes and pond volumes, plus a scoring script and a table of each tool's results.

Built with what exists

Drainage reports and review letters from county permit portals; a current AI model given each town's drainage manual, as in Enveda's test of Claude on drug-making records (see the Note "Seven more flagged"); hydro-tools to recompute runoff and pond volumes; Python for scoring.

Credit · unclaimed

Open · Planted and porous surfaces

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.

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.

Credit · unclaimed

Where it sits