{"success":true,"data":{"author":"LINKS Foundation (Montello, Arnaudo, Rossi)","bucketAccessType":"public","bucketName":"links-ads/mmflood","bucketPrefix":"","category":"Earth Observation & Geospatial","cloudProvider":"huggingface","createdAt":"2026-11-14T14:15:15.934Z","description":"Multimodal dataset for flood delineation from Sentinel-1 synthetic aperture radar, built from Copernicus Emergency Management Service (EMS) flood activations worldwide. Each activation area (folders named by EMS code, for example EMSR548-0) is tiled into GeoTIFFs with four aligned modalities: Sentinel-1 VV-VH backscatter (s1_raw), a digital elevation model from MapZen/TileZen, an OpenStreetMap hydrography map where available, and the ground-truth flood mask rasterised from EMS polygons. An activations.json file holds per-activation metadata and the train, validation and test split used in the paper. The listing (measured 2026-11-14) shows the dataset packed as eleven activations.tar.*.gz.part pieces (about 11.2 GB) that are concatenated and extracted before use. Open the tiles with rasterio or xarray. Used for SAR flood segmentation, water mapping and multimodal fusion with elevation and hydrography. The Zenodo record from the authors releases MMFlood under CC-BY-4.0; cite Montello, Arnaudo and Rossi.","displayTitle":"MMFlood: Multimodal Dataset for Flood Delineation from Satellite Imagery","doi":"10.5281/zenodo.6534637","id":"2c1e3d5e-00a7-53ef-913a-ce21a0322e2e","license":"CC-BY-4.0","licenseEvidenceUrl":"https://zenodo.org/records/6534637","objectCountEstimate":15,"probeStatus":"ok","referenceUrl":"https://zenodo.org/records/6534637","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"links-ads/mmflood\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"links-ads/mmflood\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=links-ads, license:mit, size_categories:1K<n<10K, downloads=196"],"format":"mixed","generatedAt":"2026-11-14T14:15:22.196Z","layout":{"example":"activations.tar.000.gz.part","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"activations.tar.000.gz.part","truncatedAt":512}],"sampleObjects":["activations.json","activations.tar.000.gz.part","activations.tar.001.gz.part","activations.tar.002.gz.part","activations.tar.003.gz.part"],"sidecars":[{"bytes":"---\nlicense: mit\ntask_categories:\n  - image-segmentation\nlanguage:\n  - en\ntags:\n  - semantic segmentation\n  - remote sensing\n  - sentinel-1\n  - flood\npretty_name: MMFlood\nsize_categories:\n  - 1K<n<10K\n---\n# MMFlood\nA Multimodal Dataset for Flood Delineation from Satellite Imagery.\n\n![preview](resources/preview.png)\n\n## Download\n\nThe dataset has been compacted into tarfiles and zipped, you will need to recompose it before working with it:\n\n```bash\n# clone the repository\n$ git clone git@hf.co:datasets/links-ads/mmflood\n# rebuild and extract the files\n$ cat activations.tar.*.gz.part > activations.tar.gz\n$ tar -xvzf activations.tar.gz\n```\n\n## Structure\nThe dataset is organized in directories, with a JSON file providing metadata and other information such as the split configuration we selected. Its internal structure is as follows:\n\n```\nactivations/\n├─ EMSR107-1/\n├─ .../\n├─ EMSR548-0/\n│  ├─ DEM/\n│  │  ├─ EMSR548-0-0.tif\n│  │  ├─ EMSR548-0-1.tif\n│  │  ├─ ...\n│  ├─ hydro/\n│  │  ├─ EMSR548-0-0.tif\n│  │  ├─ EMSR548-0-1.tif\n│  │  ├─ ...\n│  ├─ mask/\n│  │  ├─ EMSR548-0-0.tif\n│  │  ├─ EMSR548-0-1.tif\n│  │  ├─ ...\n│  ├─ s1_raw/\n│  │  ├─ EMSR548-0-0.tif\n│  │  ├─ EMSR548-0-1.tif\n│  │  ├─ ...\nactivations.json\n```\n\nEach folder is named after the Copernicus EMS code it refers to. Since most of them actually contain more than one area, an incremental counter is added to the name, e.g., `EMSR458-0`, `EMSR458-1` and so on.\nInside each EMSR folder there are four subfolders containing every available modality and the ground truth, in GeoTIFF format:\n\n- DEM: contains the Digital Elevation Model\n- hydro: contains the hydrography map for that region, if present\n- s1_raw: contains the Sentinel-1 image in VV-VH format\n- mask: contains the flood map, rasterized from EMS polygons\n\nEvery EMSR subregion contains a variable number of tiles. However, for the same area, each modality always contains the same amount of files with the same name. Names have the following format: `<emsr_code>-<emsr_region>_<tile_count>`. For different reasons (retrieval, storage), areas larger than 2500x2500 pixels were divided in large tiles.\n\n> **Note:** Every modality is guaranteed to contain at least one image, except for the hydrography that may be missing.\n\nLast, the `activations.json` contains informations about each EMS activation, as extracted from the Copernicus Rapid Mapping site, as such:\n\n```json\n{\n    \"EMSR107\": {\n        ...\n    },\n    \"EMSR548\": {\n        \"title\": \"Flood in Eastern Sicily, Italy\",\n        \"type\": \"Flood\",\n        \"country\": \"Italy\",\n        \"start\": \"2021-10-27T11:31:00\",\n        \"end\": \"2021-10-28T12:35:19\",\n        \"lat\": 37.435056244442684,\n        \"lon\": 14.954437192250033,\n        \"subset\": \"test\",\n        \"delineations\": [\n            \"EMSR548_AOI01_DEL_PRODUCT_r1_VECTORS_v1_vector.zip\"\n        ]\n    },\n}\n```\n\n## Data specifications\n|Image      | Description                       | Format    | Bands\n|S1 raw\t    | Sentinel-1 (IW GRD)               | GeoTIFF   | Float32 0: VV, 1: VH\n|DEM\t    | MapZen Digital Elevation Model\t| GeoTIFF   | Float32 0: elevation\n|Hydrogr.\t| Permanent water basins, OSM\t    | GeoTIFF   | Uint8 0: hydro\n|Mask\t    | Ground truth label, CEMS\t        | GeoTIFF   | Uint8 0: gt\n\n### Image metadata\n\nEvery image also contains the following contextual information, as GDAL metadata tags:\n\n```xml\n<GDALMetadata>\n<Item name=\"acquisition_date\">2021-10-31T16:56:28</Item>\n  <Item name=\"code\">EMSR548-0</Item>\n  <Item name=\"country\">Italy</Item>\n  <Item name=\"event_date\">2021-10-27T11:31:00</Item>\n</GDALMetadata>\n```\n\n- `acquisition_date` refers to the acquisition timestamp of the Sentinel-1 image\n- `event_date` refers to official event start date reported by Copernicus EMS\n\n## Run experiments\n\nYou can find the associated code in the following repository:\n```console\ngit clone git@github.com:edornd/mmflood.git && cd mmflood\npython3 -m venv .venv\npip install -r requirements.txt\n```\n\nEverything goes thro","encoding":"utf8","key":"README.md","name":"README.md","size":5379,"truncated":true}],"sourceObjectKey":"activations.tar.000.gz.part","starterCell":"from huggingface_hub import hf_hub_download\npath = hf_hub_download(repo_id=\"links-ads/mmflood\", filename=\"activations.tar.000.gz.part\", repo_type=\"dataset\")\nprint(path)","starterSamplePath":"activations.json","totalObjects":15},"summary":"Multimodal dataset for flood delineation from Sentinel-1 synthetic aperture radar, built from Copernicus Emergency Management Service (EMS) flood activations worldwide.","title":"MMFlood: Multimodal Dataset for Flood Delineation from Satellite Imagery","updatedAt":"2026-11-14T17:49:58.608Z","visibility":"public","websiteUrl":"https://huggingface.co/datasets/links-ads/mmflood"}}