We are doing science for policy
The Joint Research Centre (JRC) is the European Commission's science and knowledge service which employs scientists to carry out research in order to provide independent scientific advice and support to EU policy.
European forests provide a set of fundamental services that contribute to climate change mitigation and human well-being. At the same time, forests are vulnerable systems because the long life-span of trees limits the possibility of rapid adaptation to drastic environmental changes. Climate-driven disturbances in forests, such as fires, windstorms and insect outbreaks, are expected to rise drastically under global warming. As a result, key forest services, such as carbon sequestration and supply of wood materials, could be seriously affected in the near future. Despite the relevance and urgency of the issue, little is known about the vulnerability of European forests to multiple climate-related hazards and the adaptation benefits of alternative forest management strategies. To fill this knowledge gap we investigated the susceptibility of European forests when exposed to a given natural disturbance under different forest management scenarios. For this purpose, we assessed forest vulnerability by integrating in a data-driven framework satellite observations, national forest inventories, land surface climatic data and records of disturbances over the 2000-2017 period. The integration of these data streams is meant to capture the key drivers of vulnerability and to quantify, for the first time, the vulnerability of European forests to fires, windstorms and insect outbreaks in a systematic and spatially explicit manner. We point out that, the term vulnerability is used in this study to express to what degree a forest ecosystem is affected when exposed to a given disturbance. In order to derive risk estimates, vulnerability estimates should be integrated with hazard and exposure components, according to typical impact assessment frameworks. Results of this analyses show that in average at Europe level forest vulnerability to windstorms appears the disturbance with larger biomass loss both in relative and absolute terms (~38%, ~17 t ha-1) compared to fires (~24%, ~12.5 t ha-1) and insect outbreaks (~21%, ~9 t ha-1). Substantial spatial variations in vulnerability emerge and depict generally higher values in norther and Mediterranean regions. Overall, forest structural properties play a larger control on the vulnerability of European forests to natural disturbances compared to climate and landscape features. However, increases in temperature and changes in precipitation patterns occurred over the last two decades, have contributed substantially to make European forests more vulnerable to natural disturbances. We found that these changes in climate led to a limited increase in vulnerability at Europe for fires and windstorms and to a strong increase for insect outbreaks. However, contrasting regional trends emerging over Europe mask relevant temporal changes in vulnerability occurring at local scale. When analyses of single disturbances are combined together, results show that large part of the European forests are substantially vulnerable to at least one natural disturbance and that many of the areas more vulnerable have been subject to an amplification of vulnerability over the observational period due to changes in climate. Reducing tree age and tree density appear effective forest management strategies to reduce the vulnerability of European forests to climate-driven disturbances. The magnitude of the potential benefits appears strongly dependent on local environmental conditions. Previous assessments of future climate risks to European forests, based on catalogues of disturbances collected at country level, have showed that damage from fires, windstorms and insect outbreaks is likely to increase further in coming decades. Such intensification could offset the impact of land-based strategies aiming to increase the forest carbon sink. However, the country scale approach used in such studies do not allow to explore in detail the underlying physical processes and to elaborate adaptation strategies at appropriate local scales. It is therefore fundamental to elaborate new modelling approaches that address in explicit manner the high spatial and temporal variability of forest disturbances. In this respect, machine learning approaches and the increasing availability of multi-platform satellite observations of land surface in combination with high regional climate model simulations, represent valuable opportunities to appraise the impact of forest disturbances at a spatial and temporal resolution relevant for forest management strategies. This explorative study represents a first step towards such integrated framework.