Land & Carbon Lab

Integrated Disturbance Alerts

Integrated alerts from multiple alerting systems for monitoring vegetation disturbances in near-real-time.

Created
Aug 18, 2026
Last Updated
Nov 20, 2026
Temporal coverage
2019 -
Caution:
  • These alerts detect vegetation, forest or tree cover disturbances. This product does not distinguish between human-caused and other disturbance types. Where alerts are detected within plantation forests (more likely to happen in the GLAD-L and DIST-ALERT systems), alerts may indicate timber harvesting operations, without a conversion to a non-forest land use.  

  • Some of these systems use the term “deforestation alerts” because these detect potential deforestation events, and alerts could be further investigated to determine this.  

  • We do not recommend using deforestation alerts for global or regional trend assessment, nor for area estimates. Rather, we recommend using the annual tree cover loss data for a more accurate comparison of the trends in forest change over time, and for area estimates. Recent alerts will include false positives that have yet to raise their confidence level and may eventually be removed. Past alerts may have been removed in error from the database if rapid canopy closure precedes the additional unobscured satellite observations within 6 months. In fact, alerts can be removed in as little as 16 days in the case of GLAD-L  alerts which are removed from the dataset after four consecutive observations or more than 180 days if they are not classified as high confidence. In this case, four consecutive observations can occur in 16 days if the pixel is in an area where the satellite paths overlap. Additionally, updates to the methodologies, differing number of systems (in the case of the integrated alerts), and variation in cloud cover between months and years pose additional risks to using deforestation alerts for inter/intra-annual comparison.  

  • Alert numbers can appear to jump suddenly after a cloudy period, in the case of the optical systems which are unable to pick up new alerts where clouds or smoke obscure the images. 

  • The alerts can be ‘curated’ to identify those alerts of interest to a user, such as those alerts which are likely to be deforestation and might be prioritized for action. A user can do this by overlaying other contextual datasets, such as protected areas, or planted trees. The non-curated data are provided here in order that users can define their own prioritization approaches. Curated alert locations are provided in the Places to Watch data layer.  

  • The four alert systems have different definitions of forest/tree cover, and forest/tree cover disturbances:  

    • DIST-ALERTS: Although the system operates in all vegetation, users can toggle to view alerts within “tree cover” which is defined as all vegetation greater than 3 meters in height (2020) with greater than 30% canopy cover (2010), and may take the form of natural forests or plantations. Annual tree cover loss after 2021 is masked out.   

    • Due to the differences in baseline or forest mask in the products, it is possible that alerts outside the forest mask come from one of the forest disturbance alerting systems (GLAD-L, GLAD-S2, or RADD). 

    • GLAD-L: alerts are within “tree cover” which is defined as all vegetation greater than 5 meters in height, and may take the form of natural forests or plantations. “Tree cover loss” indicates the canopy removal of at least half a pixel and can be due to a variety of factors, including mechanical harvesting, fire, disease, or storm damage. As such, “loss” does not equate to deforestation.  

    • GLAD-S2: alerts are within the primary forest mask of Turubanova et al (2018) in the Amazon river basin, with 2001-present forest loss from Hansen et al. (2013) removed.   

    • RADD: alerts are within primary humid forests. Forest loss is defined as complete or partial removal of tree cover within a pixel, and a minimum-mapping unit of 0.1 ha is used (equivalent to 10 Sentinel-1 pixels). 

  • The input alert systems do not have the same spatial and temporal coverage:  

    • DIST-ALERT is global, and although it has been operating since January 2023, Global Nature Watch has data available since December 2023 (a few alerts may be visible in the system between January and November 2023). 

    • GLAD-L: Operating in the entire tropics (30°N to 30°S) from January 1, 2018 to the present, and from 2015 to the present (although paused for a period during 2022) for select countries in the Amazon, Congo Basin, and insular Southeast Asia. Due to a re-processing effort of Landsat imagery, the available collection of GLAD-L alerts spans from Jan 1, 2021 to the present. The alert coverage area (mask) is defined by the presence of tree cover as of the year 2000, with a canopy cover threshold of ≥10%, as well as areas of tree cover gain (per Hansen et al. 2013). Areas that experienced tree cover loss from 2014 to 2021—based on the latest available loss data at the time of the method update—are excluded from the mask. 

    • GLAD-S2: Operating in the primary humid tropical forest areas of South America from January 2019 to the present. Areas that experienced tree cover loss (Hansen et al. 2013) from 2001 to 2019 are excluded. 

    • RADD: Operating in the primary humid tropical forest areas of South America, sub-Saharan Africa and Southeast Asia with coverage from January 2019 to the present for Africa and January 2020 to the present for South America, Central America, and Southeast Asia. Forest disturbances are mapped only within the primary humid tropical forest mask from Turubanova et al. (2018) with annual (Africa: 2001 - 2018; Other geographies: 2001-2019) forest loss (Hansen et al. 2013) and mangroves (Bunting et al. 2018) removed. 

  • In order to integrate the four alerting systems on a common grid, DIST-ALERT and GLAD-L are resampled from a 30 m spatial resolution to 10 m to match GLAD-S2 and RADD. As a result, a single 30 m DIST-ALERT or GLAD-L pixel will become multiple 10 m pixels in the integrated layer. Users should use caution when comparing the analysis results of individual systems to the integrated alert layer, as the number of integrated alerts will be much greater than the number of native DIST-ALERT or GLAD-L alerts. In addition, pixels in the integrated layer may not exactly align on the map with pixels in the individual DIST-ALERT or GLAD-L layer as a result of this resampling. Due to the resampling of the 30-meter datasets to 10 meters, including the tree cover filter datasets, some DIST-ALERT pixels at the edges of tree cover may not appear in the integrated layer and analysis. 

  • Each pixel in the integrated alert layer displays the earliest date of detection reported by any contributing alerting system within a 180-day window. When multiple alerting systems detect a disturbance in the same pixel within this period, it is treated as a single event and receives the highest confidence label, and the earliest detection date is preserved. With the addition of the DIST-ALERT system, alert dates can be overwritten if a more recent disturbance is detected more than 180 days after the earlier alert. Therefore alert dates and visualizations of the integrated layer may differ from those of the individual alert system layers. 

  • The “Highest confidence: detected by multiple alert systems” level can only be achieved in the integrated alert layer, in areas and for time periods where more than one alert system was in operation for that region.  

  • Each system has its own method of determining confidence.  

    • For DIST-ALERTS, the system determines confidence level by the number of anomalous observations, with more observations meaning a higher confidence level. That is, two to three anomalies detected result in a low confidence alert, whereas four or more mean a high confidence alert. Low confidence alerts are currently not available on GFW for DIST-ALERT. 

    • For GLAD-L alerts, every new alert starts out as "low confidence" when loss is first detected (e.g. one anomalous result is detected). Alerts are then classified as high confidence when forest loss has also been identified at that location in a second satellite image within four additional (5 total) cloud-free observations. 

    • For GLAD-S2, it's the same process as GLAD-L, except alerts are classified as high confidence when forest loss has also been identified in a second satellite observation within three additional (4 total) cloud-free images.  

    • For RADD, researchers use 2 years of data to create historical image metrics showing previous forest condition, preprocess every new Sentinel-1 image, and apply a forest disturbance detection algorithm which calculates the probability that a pixel is disturbed. If the probability of disturbance is greater than 0.85, it becomes a low confidence alert. Subsequent observations within the next 90 days are used to update the probability that the forest was disturbed. When the probability reaches above 0.975, the alert becomes classified as high confidence. 

    • The confidence level may change retroactively as source data is updated. GLAD-L and GLAD-S2 alerts that have not become high confidence within 180 days are removed from the dataset.  The RADD alert system removes low confidence alerts after 90 days.

    • DIST-ALERT has the ability to detect repeated alerts in the same pixel location, unlike the other alert systems (GLAD-L, GLAD-S2, RADD) where once an alert pixel reaches high confidence, forest loss will not be detected by the same alert system at that location again. However, pixel locations where low confidence (GLAD-L, GLAD-S2, RADD) alerts have been removed from the database are subject to being alerted again by the same system. 

  • Accuracies vary across the coverage of the integrated alerts, due to different characteristics of the three alert systems – Radar (RADD) alerts for example may have more false detections in swamp forests due to the high sensitivity of short wavelength C-band radar to moisture variation. The optical systems (DIST-ALERT, GLAD-L, GLAD-S2) are more susceptible to errors from cloud interference.

WRI Data
2 Data Files