16-12-2022 - Resilience Programme Policy Brief 2022
This policy brief was funded through ForestValue Eranet Cofund Action I-Maestro (Horizon 2020 grant N° 773324) and national funding (FNR grant N° 2219NR189).

I-Maestro | Policy Brief 20222
1. Introduction
The future of our forests is uncertain, especially with regards to disturbances, such as windstorms, wildfires and bark beetles increasing due to climate change. If we want to ensure that forests continue to provide their valuable ecosystem services in challenging times, we need to investigate climate change and disturbances
effects and project how they will impact our forests in the future.
2. Disturbance risks evolution in European forests
Our analysis of disturbance trends showed that damages by all forest disturbance drivers (e.g. storms, fire, bark beetles) clearly increased from 1950 to 2019 (Figure 1). We can state this based on a comprehensive update of the Database on Forest Disturbances in Europe (DFDE) that we completed in I-Maestro. This update resulted in a unique collection of ground-based, empirical observations of forest disturbances in Europe, counting more than 170,000 records. The disturbance increase greatly impacts European forests and the services they provide to society,
especially on the local scale. The overall largest damage – 46% of total damage – was caused by wind disturbance, which generally shows large variation from year to year, while bark beetles and other biotic agents had the fastest increase recently. Whereas drought also impacted forests in the last decade, it could not be distinguished as a separate disturbance cause, as it often acts as a predisposing factor triggering another disturbance cause such as bark beetles or fire, which is then recorded in the statistics.
2.1 Predicting the risk of drought-induced mortality in Poland
We used the data we collected from Scots pine forests in Poland to predict the risk of mortality caused by drought. Furthermore, we investigated the effects of stand structure and site productivity. We found that Scots pine mortality was mainly driven by a water deficit in the period May-August of the previous year.
Moreover, we documented that the oldest stands growing on the most productive sites had the highest susceptibility to increased mortality during drought, which could motivate forest managers to shorten the rotation period (= harvest earlier) 2.2 Forest recovery after disturbances Within I-Maestro, we compiled a tree
regeneration dataset from ground-based inventories to analyse patterns and drivers of forest recovery after disturbances that caused high tree mortality over large, forested areas.
The dataset covers 30 disturbance events and more than 100 forest sites, both managed and unmanaged, over the entire temperate zone of Europe (Figure 2). Overall, our analyses showed that after large and severe disturbances
temperate forests tend to recover their structure better than their species composition. Their recovery was less successful after fire than after other disturbances. Furthermore, we found that forest regeneration is affected by environmental factors, such as elevation and incident light radiation, indicating that mountain forests were recovering at a slower pace than those at lower
elevation and a milder climate. Another relevant observation is that post-windthrow salvage logging had a negative impact on regeneration
compared to disturbed areas that were left unmanaged. Interestingly, planting after logging did not show evidence of positive effects on post-windthrow regeneration.
3. Role and contribution of forest simulation modelling to make our forests more resilient

Forest simulation models help us to explore our decision space for the future. However, scientific analyses using models are often too complex to directly inform decision making of e.g. forest owners. Results of forest simulation studies
aim to inform advisors who can then translate results into practical guidance: What species will be suitable under the projected future climate conditions? What is the range of impacts under low vs high climate change scenarios?
Where should we focus disturbance prevention measures?
The modelling studies can 1) raise awareness about the ongoing changes in climate and disturbance regimes; 2) explain how the intensified climatic extremes and disturbances differ from the past and what impacts we might face when continuing with “business as usual”; and 3) provide adaptive response options for
practitioners to enhance the resilience of forests.
3.1 Generating case study data for modelling
In I-Maestro we set up three case study landscapes in France, Poland and Slovenia to assess forest dynamics under different climate change, disturbances and management scenarios. These landscapes were used to evaluate the effect
of complexity on the resilience of forests to hazards like windstorms, fires, or insect attacks.
By complexity, we mean the local organization of trees, from homogeneous to heterogeneous species and sizes, or the arrangement of stands at larger scales. To address this question, we needed characteristics of every tree in the landscape, covering thousands of hectares.
We developed a method to derive the tree characteristics for very large areas with good accuracy, based on Airborne Laser Scanning (ALS). ALS provides data for the entire landscape and can be combined with National Forest Inventories for both the calibration of the ALS and the downscaling to the tree level. For the three case studies of I-Maestro, we generated about 42.4 million trees from 51 different species.
3.2 Using empirical disturbance data to set up disturbance scenarios.
Based on the disturbance data collected in the Database on Forest Disturbances in Europe we developed a new approach to create disturbance scenarios. By using the expanded records of empirical observations of wind and fire disturbances, it was possible to generate disturbance scenarios which represent a realistic year to year variability of regional disturbance risks. We then applied these scenarios in forest simulation models to explore forest dynamics under disturbances. We did this because we consider disturbances even more important for assessing the future of our forests than gradual changes in temperature and precipitation
(rainfall and snow). Our analysis in I-Maestro stressed that it is very important to validate the projected disturbance damages by using regional disturbance data (e.g. provided by forest inventories).
3.3 Trade-off between diversity at different spatial scales
We analysed the effect of diversity at the local scale (diversity of trees in a stand), as well as at the landscape scale (diversity of stands in the landscape). We found that there are trade- offs and synergies between these two levels of diversity, depending on the case studies: The mountainous landscape in the Bauges
(France) is an example of a trade-off between stand and landscape scale diversity. It contains mostly uneven-aged stands with high local diversity, but these stands are almost all identical: the stand level diversity is high, but the landscape diversity is quite low. Increasing landscape diversity by adding even-aged stands reduces the stand scale diversity. Conversely, in plain forests such as the Milicz case study,
landscape and stand level diversity interact synergistically: adding uneven-aged stands in this landscape mostly composed of even-aged stands increases landscape diversity and simultaneously also enhances the local diversity.
Authors: Gesche Schifferdecker, Gabriela Rueda, Marcus Lindner & I-Maestro Project Team