Landslide scar inventory from the February 2026 rainfall event in Juiz de Fora (Minas Gerais, Brazil)

DescriptionThis dataset contains a polygon inventory of landslide scars mapped in the municipality of Juiz de Fora (Minas Gerais, Brazil), associated with a rainfall-triggered event in February 2026.MethodologyLandslide scars were mapped through manual visual interpretation in a GIS environment using PlanetScope satellite imagery (~3 m spatial resolution). Post-event images acquired on 2026-03-02 and 2026-03-08 were compared with a pre-event image from 2026-01-12 to identify new landslide features. The images were accessed via the Brasil Mais platform (© Planet Labs Inc.) and analyzed using the Planet plugin in QGIS.The inventory focused on urban areas and sectors previously recognized as susceptible to landslides or with documented occurrences. Landslide scars were delineated as polygons, including source areas, transport zones, and associated deposits.A total of 283 landslide scars were identified.Event descriptionThe landslide event was associated with intense and recurrent rainfall episodes affecting the Juiz de Fora region. During the most critical phase (2026-02-23 to 2026-02-29), rainfall accumulations frequently ranged between 150–200 mm, with areas reaching 200–250 mm and locally exceeding 300 mm. These conditions indicate repeated rainfall events acting over a previously saturated system, contributing to widespread slope instability.Data descriptionGeometry: PolygonsFormat: GeoPackage (.gpkg)Coordinate system: WGS84 EPSG4326Number of features: 283LimitationsSmall landslides, particularly those associated with cut-and-fill slopes in densely urbanized areas, may not be detectable due to image resolution (~3 m). Mapping uncertainties are inherent to visual interpretation, and vegetation or urban structures may obscure landslide features. Therefore, the inventory is representative of landslides with larger spatial expression.Usage notesThis dataset is intended for research and educational purposes. Users should consider mapping uncertainties and resolution constraints when interpreting the data.CreatorsRodrigo Augusto StabileCemaden (National Center for Monitoring and Early Warning of Natural Disasters/Brazil)Contact: [email protected]  Keywordslandslides; landslide inventory; rainfall-triggered landslides; mass movements; geomorphology; Brazil; Juiz de ForaLicenseCreative Commons Attribution 4.0