BRENNEN

Industries / Energy, water and waste / Environmental services

Environmental remediation

Environmental remediation finds pollution in soil and groundwater and then removes, destroys or contains it, so that the land and water can be used safely again.

1980

In Lekkerkerk, near Rotterdam in the Netherlands, chemicals leaking from hundreds of drums buried under a new housing estate were found in the groundwater, and on 2 May 1980 the government ordered the estate emptied by 1 June so that the polluted ground could be dug out.

parlement.comcsmonitor.comrivm.nl

37

Milestones achieved across 6 domains.

7

To watch for: announced, not yet confirmed outside.

12

Open deliverables. Nobody has shipped them.

In one minute

Where it started

Cleaning up polluted land became an industry around 1980, after waste buried under homes came to light at Love Canal in New York and at Lekkerkerk in the Netherlands. The US Congress set up its Superfund for cleanups in 1980, and by January 1981 the Dutch government counted about 3,000 chemical waste dumps in the country, 300 of them a serious threat to health. The first answer was to dig the pollution out, but by the early 1990s the Dutch found that too costly for every site and began cleaning land only as far as its intended use required.

Where it stands

In the US, EPA counts 583,313 confirmed leaks from underground fuel tanks; 530,454 had been cleaned up and 52,859 were still waiting in March 2026. At US Superfund sites, digging polluted soil up and trucking it away is still the most common choice for soil, while polluted groundwater is now treated in place underground more often than it is pumped out. Worldwide, the UN's FAO and UNEP reported in 2021 that there are still "significant knowledge gaps and uncertainty about the number and extent of areas affected."

What is open

Much of what is known about cleanups sits in reports that consultants file with regulators, one site at a time. Open jobs turn those reports into checked public data: how in-place treatments begun since 2012 have worked, how fast PFAS break through carbon and resin filters, how long injected carbon barriers hold, and how often closed sites are reopened. Others build shared tests for AI that reviews Phase I site assessments, for methods that map plumes and forecast when a site will be clean, and for checks from satellites and aerial images.

Domains

AchievedWatch forOpen

Advances to build with

3 research notes

Open in environmental remediation

3 of 12 open deliverables

Open · Site investigation

Score AI on Phase I

A public test set of Phase I environmental site assessments, each with the public records the assessor used and the "recognized environmental conditions" (RECs, signs that a property is or may be polluted) the assessor found, plus a scored run of an AI agent given the same records.

Built with what exists

Public Phase I reports from city and state brownfield programs; EPA's public facility and cleanup-site databases; the Library of Congress's Sanborn fire-insurance maps and USGS historical topographic maps; a model built for long documents, such as Muse Ember (see the Note "The jobs agents repeat"); Gemini Decision 1 to answer a fixed yes-or-no question for each record (see the Note "A thousand questions, one pass"); an agent that cites the record behind each finding (see the Note "Findings that cite notes"); open agents such as matrix-esa-agent as a baseline.

Credit · unclaimed

Open · Removing and treating soil

Track polluted soil after 2012

A public yearly series, 2006 to 2025, of the polluted soil that England's permitted waste sites received, split into landfill and treatment or recovery, built from the Environment Agency's Waste Data Interrogator. Soil is coded in the European Waste Catalogue as 17 05 03* (soil holding hazardous substances) and 17 05 04 (other soil). The series then measures the change after April 2012 against HMRC's forecast.

Built with what exists

The yearly Waste Data Interrogator files on data.gov.uk (2006 to 2025, Excel extracts of waste received at each permitted site); HMRC's landfill tax statistics; Python with pandas or DuckDB; a coding agent that writes and tests a parser for each year's file layout, such as one built on Inkling 2 Small (see the Note "Checked without being asked").

Credit · unclaimed

Open · Pumping and treating groundwater

Rerun the 2004 pumping contest

Port the three 2004 pump-and-treat optimization problems (Umatilla, Tooele and Blaine) from their original MODFLOW96 and MT3DMS files to current open tools, MODFLOW 6 and FloPy, with a scoring script that checks every limit on pumping and concentration, and publish them as an open benchmark. Then test modern optimizers and an AI coding agent against the best 2004 answers.

Built with what exists

The ESTCP files on frtr.gov; MODFLOW 6 and FloPy from the US Geological Survey; open optimizers such as SciPy, pymoo or Nevergrad; a coding agent to convert the files and write tests, such as one built on Inkling 2 Small (see the Note "Checked without being asked").

Credit · unclaimed

Where it sits