Determining the spatio-temporal patterns of rockfalls, such as the zonation of hazards and the assessment of associated risks, can be challenging due to poor historical archives. Dendrogeomorphological methods cover this lack of data and provide reliable reconstructions of rockfall activities over several centuries. These methods are based on the signals recorded in the tree rings that are affected by the mechanical impact of falling rock fragments. In this study, we analyzed the spatial and temporal distribution of rockfalls in a 0.19 ha forest area in the Southern Carpathians. We collected 170 samples (100 increment cores and 70 stem discs) from all 40 Picea abies (L.) Karst trees identified in the study area (1 tree/47 m²). This allowed us to date 945 events between 1817 and 2021, which we then compared with available weather records. Our results show the main trajectory of falling rock fragments from the source area, as well as significant temporal variations in process activity. These variations correlate only slightly with fluctuations in meteorological parameters. Despite the expected intensification of natural hazards due to climate warming, our study area shows a general trend towards a slight decrease in rockfall activity at present.

Rockfalls are a destructive geo-hazard in Romania's steep environments that pose significant material damage and loss of life, even when involving small rock volumes, owing to their high mobility and unpredictability. This study presents the first national-scale assessment of rockfall potential occurrence in Romania, applying a multi-criteria statistical and geospatial analysis based on the Weight of Evidence method. By integrating 1,743 historical rockfall sites and 19 environmental variables as driving factors, we mapped and statistically assessed the potential rockfall distribution of their occurrence across the Carpathian and Subcarpathian regions. Findings reveal that over 18,000 km² of Romanian territory is suitable for rockfall assessment, while approximately 1,500 km² face high rockfall occurrence potential. The Southern Carpathians emerged as the most vulnerable area, and all over the Carpathians and Subcarpathians statistical analysis highlighted that topographical and environmental factors are the primary drivers of rockfall distribution. Our model achieved 94% predictive accuracy, confirming the reliability of the resulting potential rockfall occurrence map, as well as its utility as a decision-support tool. These findings provide a robust scientific reference for national and local authorities to mitigate rockfall geohazards, manage rockfall risks and improve infrastructure resilience in Romania.