Skip to content
Peta Satu Ruang
All data

Floods

See where satellites have seen floods through 2026 in every city and regency, how much near homes, including the last few days, from Sentinel-1 radar that sees through cloud.

Sources: Copernicus GFM (Sentinel-1)

Flooded area from satellite

Download data

Choose a file format. Downloads do not follow the current table search or page. Source and licence details are included; credit both when using or sharing data.

Floods per day

One row for every city/regency and day a satellite saw it: flooded area, the part near homes, area seen by radar, and the number of passes. Map points are the city/regency centre. A day without a row was not seen, which does not mean it was dry.

API

Responses use JSON and include stored data with its sources. Outside this site, the API needs an API key; downloads stay free. Credit the source and its licence.

Send your key in the X-API-Key header when accessing the API from an app or script.

  • GET /api/floods

    Largest flooded area in one day per city/regency and year since 2026 from Copernicus GFM, with the part near homes and the days seen by satellite, and every flooded day of the last 30 days. With tahun=YYYY, one year per month.

  • GET /api/export/banjir?format=xlsx|csv|geojson|shp&tahun=2026&area=32

    Every stored day of one year: one row per city/regency and day a satellite saw it. tahun: without it, the newest year. area: a province or city/regency code, optional.

Methodology

Data coverage

The flood map and the map of areas that cannot be read from every Sentinel-1 pass over Indonesia since 1 January 2026. A new pass usually appears about 3 hours after the satellite flew over, and we check every 15 minutes.

Processing

  • How GFM maps floods: the Sentinel-1 radar sends microwaves that pass through cloud and work day and night. Calm water reflects them away, so it looks dark. Three teams (DLR, TU Wien, LIST) each look for 20 m cells darker than usual, and GFM uses what they agree on. Permanent water such as the sea, lakes and large rivers is not counted as flood.
  • How we count it: each 300 × 300 km map sheet is read at 80 m cells. Each flooded cell is measured on the earth's surface and goes to the city or regency that holds its centre, using BIG boundaries. The result differs by a few percent from a count of 20 m cells.
  • Area seen by radar is the part of an area the satellite passed over and could read that day, outside GFM's excluded areas. It is read at cells of about 640 m.
  • One day is a WIB day. When a satellite passed more than once, the largest area is kept, not the sum. Each year's figure is the day with the largest flooded area, with the number of days the satellite saw the area.
  • An area no pass saw that day has no figure. It is shown as no data, never as zero or safe.
  • Near homes: each 20 m flooded cell is matched with the ESA WorldCover 2021 land cover map. A cell with buildings within 200 m counts as near homes. Its area is the day's flooded area times the share of cells near homes. A new day is counted soon after it is read; until then the figure is empty, not zero.

Limitations

  • The satellites pass every few days, not daily. A flood that drains before a pass is not seen. After Sentinel-1B failed in December 2021 passes were rarer until Sentinel-1C joined in 2025, so flooded days cannot simply be compared between years.
  • Radar cannot see water between closely packed buildings or under thick forest. Floods inside cities such as Jakarta are often missed or counted much smaller.
  • Lakes that grow in the rainy season, flooded rice fields, fish ponds and swamps can read as flood, for example Lake Tempe in Wajo, the Mahakam lakes in Kutai Kartanegara, and the Karawang, Indramayu and Pinrang plains.
  • This is flooded area, not its impact. Deaths, evacuees and flooded homes are not in satellite data.
  • Near homes does not mean houses under water. Most of it is rice fields or land at the edge of villages. WorldCover's buildings are from 2021, so newer settlements are not counted.

Licence & usage

Copernicus Emergency Management Service, Global Flood Monitoring (GFM), under the CEMS terms: free to use and share with credit to the source.

Data sources and terms