- Activity
Land Use Change Alerts (LUCA)
Detect forest loss from deforestation, degradation, and fire-related disturbances
Lead scientist
Access LUCA via an open Google Earth Engine app, and as raster data by request.

Near-real-time alerts on forest disturbances, globally
LUCA (Land Use Change Alerts) is a near-real-time global forest land use change monitoring product derived from Sentinel-1 Synthetic Aperture Radar (SAR) data. It detects forest loss events such as deforestation, degradation, and fire-related disturbances by analyzing changes in radar backscatter on a rolling biweekly basis.
Designed for timeliness and weather-proof detection, LUCA provides early warning alerts with high spatial and temporal resolution for all forest types globally, making it a valuable tool for forest monitoring, law enforcement, and climate mitigation initiatives.
Data at a glance
- Spatial resolution
10 meters (m), with a 0.05-ha minimum mapping unit that can be adjusted
- Spatial coverage extent
Global
- Temporal frequency
Biweekly
- Temporal domain
2018 to present
- File format
Google Earth Engine asset, Geotiff, Cloud optimized GeoTIFF
Near-real-time forest alerts based on synthetic aperture radar
Input data
We used a Sentinel-1 SAR ground range detected (GRD) images in VV and VH polarizations in both ascending and descending orbits covering all major forest biomes globally from January 1, 2016, until the present day. The Sentinel-1 SAR images were acquired in the interferometric wideswath mode with a resolution of 5×20 m in the range and azimuth directions, respectively, and were processed for thermal noise removal, calibrated to sigma nought, and corrected for terrain using range Doppler terrain correction with a 10 m pixel size prior to being ingested into Google Earth Engine (GEE).

Modeling approach
We first preprocessed the Sentinel-1 images to the analysis-ready data (ARD) level. Second, we engineered features to define the statistical properties of stable forest, then synthesized the reference labels and sample training dataset in different forest biomes and trained a machine learning model to derive forest and forest land use change probabilities. The derived forest and forest land use change probabilities were refined in a probabilistic approach, and forest land use change alerts were generated. Once the alerts were generated, a post-processing step was applied to remove noisy alerts and provide the final alerts at a minimum mapping unit of 0.05 ha, which are displayed in the online app.
Validation approach and results
We assessed the accuracy of the generated alerts from 2018 to 2024 by using area-adjusted accuracy assessment as suggested in Olofsson et al. (2014) and Oloffson et al. (2020). These alerts were generated in an emulated near-real-time mode similar to other studies conducted in detecting near-real-time forest disturbance alerts.
We generated 450 samples in three strata using stratified random sampling in each geographic region of the world: South America, Central America, North America, Europe, Asia, Southeast Asia, Oceania, and Eurasia. We derived 150 random samples from the forest land use change strata and 300 random samples from the forest baseline and buffer area within a 200-meter distance from alert pixels. The buffer areas were sampled to target the omission errors that occur in the proximity of detected forest land use changes. In the tropics, the reference labels were extracted visually from NICFI Planet monthly mosaic images. In North America, we used National Agriculture Imagery Program (NAIP) aerial ortho-images to extract the reference labels; and for Europe and Eurasia, we used cloud-free Sentinel-2 multispectral images.
The benchmark forest cover map for 2018 was generated from the Landsat-based Hansen annual tree cover product, with forest cover changes before 2018 removed using the forest cover loss layer in the same product. Some forest change events that happened before 2018 but were not included in the forest baseline dataset were detected by LUCA as new alerts in 2018. These events were considered actual alerts, not commission error, but were removed from the accuracy assessment sampling. Samples extracted from boundary pixels of large disturbance detections that were clearly shown in the higher resolution Planet images were also discounted as commission errors.

Forest alerts for wildfire detection, risks to protected areas, and more
Wildfire Detection
In Brazil's Cerrado savanna, wildfires burned through 97% of the Serra das Araras conservation area in 2024, according to a news report in Mongabay. While fire is an important part of the Cerrado’s ecosystem, human-induced changes to fire patterns pose an existential threat to the reserve and the many species living within it. In August to December 2024, LUCA detected more than 1,000 hectares of disturbance in the reserve, an area more than twice the size of San Francisco’s Golden Gate Park.
Encroachment into Protected Areas
The Environmental Performance Index finds that the health of Nicaragua’s forests have declined by over 11% throughout the last decade. Reporting in Mongabay points to unlawful encroachment of cattle ranching, mining, and logging into the country’s protected forests.
The impacts of these activities are particularly visible in Indio Maíz and Bosawás, two large biosphere reserves located on the eastern and western borders of the country. In both reserves, forest disturbance alerts in 2024 far surpassed that of 2023, according to LUCA data. In Bosawás, total disturbance alerts in the second quarter of 2024 reached 34,943 total hectares–a nearly 68% increase from the same timeframe in 2023. In Indio Maíz, the 1,323 hectares of disturbance alerts in April 2024 alone were roughly equal to total disturbance alerts for all of 2023.
Shifting Cultivation
Shifting cultivation is an agricultural practice in which plots of land are temporarily cultivated, abandoned, and left to regenerate during a fallow period. In the last three decades, shifting cultivation has become the leading cause of forest disturbance in Laos. Population growth and the recent construction of a railroad connecting Laos with China have further increased pressure on the country’s forests.
LUCA data reveals that in Q2 2023, the area of forest disturbance in the Luang Prabang province reached 72,756 total hectares, an area roughly the size of New York City. This is the highest quarterly total for the province since LUCA data became available in 2018.
References
A. Mullissa, S. Saatchi, R. Dalagnol, T. Erickson, N. Provost, F. Osborn, A. Ashary, V. Moon, D. Melling, “LUCA: A Sentinel-1 SAR-Based Global Forest Land Use Change Alert” Remote Sensing 16 (12) (2024) 2151
Mullissa, A.; Vollrath, A.; Odongo-Braun, C.; Slagter, B.; Balling, J.; Gou, Y.; Gorelick, N.; Reiche, J. Sentinel-1 SAR Backscatter Analysis Ready Data Preparation in Google Earth Engine. Remote Sens. 2021, 13, 1954.
Mullissa, A.; Reiche, J.; Herold, M. Deep learning and automatic reference label harvesting for Sentinel-1 SAR-based rapid tropical dry forest disturbance mapping. Remote Sens. Environ. 2023, 298, 113799.
Olofsson, P.; Foody, G.M.; Herold, M.; Stehman, S.V.; Woodcock, C.E.; Wulder, M.A. Good practices for estimating area and assessing accuracy of land change. Remote Sens. Environ. 2014, 148, 42–57.
Olofsson, P.; Arévalo, P.; Espejo, A.B.; Green, C.; Lindquist, E.; McRoberts, R.E.; Sanz, M.J. Mitigating the effects of omission errors on area and area change estimates. Remote Sens. Environ. 2020, 236, 111492.
Team
Page last updated Nov 16, 2026
All datasets
Stay Connected
Subscribe to our monthly newsletter and be the first to know about new data releases, research, and events.