A satellite can track water in 800,000 km of Asian canals in a region where irrigation helps feed roughly 3 billion people

A radar satellite has been shown to read water levels along roughly 800,000 kilometers of irrigation canals across Asia, a job that ground crews have never been able to do at that scale. The University of Washington team behind the work, led by graduate research assistant Mridul Sharma, reported that the satellite measured water levels with moderate to high confidence at more than 85% of locations it examined.

The canals sit in a region where irrigation plays a critical role in feeding roughly 3 billion people, the university said on September 30, 2026. That figure describes the region’s irrigated food supply, and the release does not claim that the canals alone feed all of those people.

800,000 kilometers across 22 countries

The University of Washington release says the team focused on canals in 22 countries across Asia and looked at the satellite’s data along 800,000 kilometers of them. The canal map came from GRAIN, the Global Registry of Agricultural Irrigation Networks, which the university dates to 2025 and which a March 2026 paper in Earth System Science Data describes as identifying about 3.82 million kilometers of agricultural canals in 95 countries. Its authors, Sarath Suresh, Faisal Hossain, Vimal Mishra and Nehan Hossain, built it from OpenStreetMap data and a Random Forest classifier that separates canals from rivers and streams, and the registry also has online documentation. Sharma put the division of labor this way: “GRAIN mapped where canals are. SWOT revealed what was happening inside them.”

The satellite is SWOT, the Surface Water and Ocean Topography mission, which NASA describes as its first global survey of Earth’s surface water. NASA and France’s space agency CNES developed it jointly, with contributions from the Canadian Space Agency and the UK Space Agency. According to the university, SWOT has been in orbit since December 2022 and passes over a given place at least once every three weeks, using interferometric radar that compares scans over time to detect changes in elevation.

The release does not list the 22 countries. The work appears in Geophysical Research Letters, and the university links to the paper for the full methods.

Confidence scores and where the signal fails

Sharma overlaid SWOT’s radar data on the canal map and gave every kilometer a confidence score, based on how well the satellite’s view matched the contours of the land and how strong the signal was relative to noise. The breakdown is uneven: 37.5% of kilometers were highly observable, 46.9% moderately observable and 15.6% poorly observable. The more-than-85% figure in the release refers to locations rather than kilometers, so it should not be set directly beside the kilometer shares, which add up to 84.4% for moderate-or-better.

The weak spots are physical. Many canals are narrower than 20 meters, which is at the low end of what SWOT can observe, and the mission’s designers did not expect it to see fluctuations in small canals at all. Dense vegetation along the banks was the biggest obstacle. Confidence was highest where canals were wide and well organized, the slopes smooth and the surroundings open.

To check the scores, the team compared them with real canal measurements in the United States and found that the confidence levels matched the system’s actual abilities. The university calls the Asian analysis a proof of concept and says performance should improve as better algorithms are developed.

Farm-level uses and the people behind the study

Without satellite coverage, farmers and water managers depend on scattered gauges, inspections and local reports. The university’s release says the result could give farmers around the world a powerful new tool to manage precious water resources, prepare for droughts and improve food security in their communities, and describes one scenario: a farmer who sees from SWOT data that too little water will reach a rice field could switch to a less water-intensive crop such as corn or wheat. That example is presented as a possibility, not a result the study demonstrated.

Faisal Hossain, the University of Washington professor of civil and environmental engineering who is the paper’s senior contact, is quoted in the release saying the work may represent a fundamentally new way of managing the water conveyance systems that sustain modern agriculture. Co-authors include Sanchit Minocha and Shahzaib Khan of the university, Sarath Suresh, now at Virginia Tech, and Tamlin Pavelsky of the University of North Carolina at Chapel Hill. NASA funded the work, and Hossain is the listed contact for questions about it at the university. Suresh completed the research as a University of Washington doctoral student before moving to Virginia Tech, a detail that ties the canal registry and the satellite analysis together under one research group.

The gap that remains is the 15.6% of canal kilometers the satellite sees poorly, a share the university ties to narrow channels and dense vegetation and expects algorithm improvements to shrink.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview