Key finding
The RADD-Europe system uses Sentinel-1 C-band synthetic aperture radar, which penetrates clouds and operates day and night at 20-meter spatial resolution with a 3-to-6-day revisit time across Europe. It integrates near-real-time ERA5-Land temperature data and Copernicus forest type information to correct for freezing temperatures and seasonal phenology that affect radar backscatter. The system achieves 91.2% user accuracy and 74.5% producer accuracy, with a median detection delay of 27 days that can be reduced to 1 day through retrospective correction. It captures diverse disturbance regimes: winter harvesting in northern Europe, spring sanitation cutting for bark beetles in central Europe, and summer wildfires in southern Europe.
What the research asks
Operational forest disturbance monitoring across Europe has been limited by persistent cloud cover that blinds optical satellites for weeks at a time. Radar sensors can see through clouds, but building a continent-scale system that handles Europe's diverse disturbance regimes, from winter harvesting in Scandinavia to bark beetle outbreaks in central Europe and summer wildfires in the Mediterranean, is technically challenging. This study asked whether Sentinel-1 C-band synthetic aperture radar, integrated with temperature and forest type data, could deliver accurate near-real-time forest disturbance alerts across all of Europe with a detection delay measured in days rather than months.
What it finds
The system, called RADD-Europe, achieves a user accuracy of 91.2 percent and a producer accuracy of 74.5 percent when validated against high-resolution Planet satellite imagery. When errors from the European forest cover mask (which tends to overestimate forest height and density in some areas) are excluded, user accuracy rises to 99 percent. The median detection delay is 27 days relative to the first available high-resolution optical reference image, and the authors note that this can be reduced to 1 day through retrospective event-based correction.
The system captures the full diversity of European forest disturbance regimes. In northern Europe, it detects winter harvesting operations. In central Europe, it picks up spring sanitation cutting to control bark beetle outbreaks. In southern Europe, it detects summer wildfires. The system also generates intra-annual disturbance seasonality estimates that reveal variation in forest management practices across the continent. Notably, it improves detection of small-scale disturbances such as group fellings in Romania, which are often missed by annual optical-based products.
Why it matters
For earth observation professionals, this paper represents a significant operational advance. The RADD-Europe system is already live and publicly accessible, providing near-real-time alerts that can be integrated into national forest monitoring systems, carbon accounting frameworks, and biodiversity observation networks. The use of open-access Sentinel-1 data means the system is replicable in other cloudy regions of the world, including the tropics, where persistent cloud cover has long been the primary obstacle to operational satellite monitoring.
For supply chain monitoring and regulatory compliance, the near-real-time capability is valuable. Companies sourcing timber or forest-risk commodities from Europe can use the system to detect potential deforestation or degradation events in their supply chains within weeks rather than months. For national REDD+ programs and forest carbon project developers, the system offers an independent, transparent, and publicly available data source for verifying forest carbon claims. The alert data is available through Google Earth Engine, enabling easy integration with existing monitoring workflows and platforms.