Abstract
The intensity, frequency, and spatial distribution of forest disturbance regimes across Europe are shifting due to climate change. This raises concerns about the vulnerability of forest ecosystems and the impacts on the goods and services that they provide. Protection against natural hazards is a key service provided by forests in the Alps but current protective effects are threatened by the growing incidence of disturbance events such as windstorms, heavy snowfalls and drought. For planning effective management interventions, detailed information on the patterns of recent forest disturbances and quantifications of their impacts on protection forests are necessary. In our study we focused on a region in the Italian Alps (South Tyrol) with the aims of: i) providing a wall-to-wall disturbance map by agent type (wind, snow, beetles) and an analysis of the spatial patterns of disturbance agents and their interaction, and ii) quantifying the loss of protective effects in protection forests and areas with residual protection given by standing dead trees due to bark beetle.
We analyzed Sentinel-2 timeseries to map disturbances covering the period 2019-2023. We then applied a supervised machine learning classifier leveraging multisource predictors to attribute a disturbance agent to each disturbed patch. Afterwards, we explored the correlation between the areas disturbed by different agents and assessed the areas of protection forest which were affected by disturbances. For these areas, we performed a pixel-based classification to identify areas with residual protective effects given by standing dead trees (i.e., pixels with dead canopy but not downed or salvaged yet) due to recent bark beetle outbreaks.
Our results showed that, over a period of five years, disturbances affected 5.9% of the forests of the study area. Damages due to windthrow (1.6%) and snow (1.3%) had a comparable cumulated impact, while bark beetle caused much larger damages (3%). Snow-damaged areas correlated more strongly with bark beetle damage than wind disturbances. Notably, 5.6% of protection forests in the area were disturbed, with bark beetles causing disproportionately higher impacts compared to the other two agents. Overall, about 1.3% of protection forests still provide some level of protection because they are covered by standing dead trees. These are forests that will soon lose their protective function and should be given high priority in management planning. These findings provide the first detailed mapping of recent disturbances in the region and highlight critical areas where protection forests can no longer offer adequate hazard mitigation. By identifying forests most at risk of losing their protective function, we offer useful information for managers to plan near future interventions in these areas.