Object-based models map burned forests in Mediterranean transfer tests
Object-based machine-learning models mapped burned forests from Sentinel-2 imagery, retaining overall accuracies above 83% in transfer tests at selected Mediterranean ...
Object-based machine-learning models mapped burned forests from Sentinel-2 imagery, retaining overall accuracies above 83% in transfer tests at selected Mediterranean ...
Laboratory hyperspectral data combined with environmental and satellite covariates improved machine-learning estimates of topsoil organic carbon across Northeast China’s croplands.
A Chiba University machine-learning model reconstructed missing regions in global satellite rainfall maps with higher accuracy and image quality than ...
A physically constrained Ka-band radar method achieved a 0.092-kilometer mean absolute error in 10 locally conditioned coastal fog events, demonstrating ...
A multi-source study of 965 reclassified settlements in Uzbekistan found that administrative recognition grouped places already urban in 2005, sites ...
A drone-mounted LiDAR survey identified three previously unregistered Aboriginal rock shelters containing rock art and stone artefacts in Australia’s Blue ...
A 2026 Perspective proposes a volumetric monitoring framework to detect underground urban expansion, connect it to environmental and infrastructure evidence, ...
KAIST’s GeoCore-9B outperformed task-specific systems on key public-dataset measures for cloud removal and led some radar-to-optical conversion metrics.
A 23-year satellite analysis of northern peatlands reveals widespread local greening but no significant map-scale increase in leaf area, with ...
Gross primary productivity, or GPP, is the amount of carbon captured by plants through photosynthesis before subtracting the carbon used ...
© 2026 Scienmag - Science Magazine
© 2026 Scienmag - Science Magazine