Forests as nature-based solutions for disaster risk reduction

Abstract/Description

Over the last few decades, the increase in high-intensity forest fires that are difficult to control and the growing risk of fires in no originally fire-prone areas have given rise to concern among forest managers. In addition to the high mortality of trees, leaving a significant number of standing burnt stems, the structural damage caused by fire can vary from a reduction in the volume of the crown to the partial destruction of the root system, through to carbonisation of the wood of standing stems. On a finer scale, the mechanical properties of stems following a forest fire are poorly reported. However, the impact of heat treatment on wood properties is better known. Associated with a loss of mass, treated wood will show improved material hardness and durability over time, but also greater vulnerability to impact and a reduction in modulus of rupture, demonstrating a potential alteration in the mechanical properties of the stem. To quantify the impact of fire on the protective effect against rockfalls, current used models describe the protective effect of a forest as a function of stem density (i.e. probability of impact) and the ability of trees to dissipate energy on impact with a block (Ediss). However, these models were calibrated on healthy trees, implying homogeneous wood properties in the tested scenarios. This work therefore aims to estimate the maximum energy dissipation capacity of burnt trees in order to assess the protective effect of a forest after a fire. To do so, we proposed a new equation to compute Ediss at individual scale. The calibration is conducted through laboratory tests on a series of green and artificially burnt stems. In parallel, we conducted a case study using stems sampled in Bitsch forest (Wallis, CH), where the forest suffered an intense and uncontrollable forest fire in July 2023. Finally, the impact of the new Ediss values on the protective effect against rockfall of the forest stand was quantified using RockyFor3D software.

Abstract/Description

Protective forests are of critical importance to ensure the security of human life, infrastructure and stability of ecosystems in mountainous regions. Particularly in the Alps, natural disturbances, such as windthrows, fires and insect outbreaks, pose substantial challenges for planners and practitioners aiming at the maintenance of the protective effect of these forests. Consequently, the assessment of residual protection offered by biological legacies, including downed logs, has become a key priority to study the effectiveness of protective forests and possible solutions in the eye of climate change. In our case study, we benefit from access to several study sites where the forest managers have intentionally left specific, known amounts of the former stand as horizontal deadwood in the sites, potentially increasing surface roughness and enhancing the protective effect against rockfall. In this study, we use rockfall simulations to analyze the roughness effects of horizontal deadwood and evaluate the effects of different post-disturbance management strategies. Specifically, we determine terrain roughness values for 5 study sites in the Italian Alps, analyze and compare them between unmanaged windthrow areas in the process of ongoing natural decay and manipulated stands with varying amounts of deadwood on the ground. Our findings aim to provide standardized roughness parameters for rockfall models such as RockyFor3D, which will be applied within a single study area. Furthermore, we aim to offer practical insights that can be transferred to other areas and assist forest practitioners and decision-makers in developing effective protection strategies for protective forests following disturbance events. This research adopts a multiscale methodology, employing diverse techniques and data sources such as field studies, lidar- and photogrammetric UAV data. The objective of this study is not only to evaluate the protective effect of the biological legacies under the given scenarios, but also to refine and enhance methodological approaches for assessing the protective role of horizontal deadwood in mountain forests.

Abstract/Description

Forests can have a braking effect on moving snow avalanches, contributing to avalanche protection, a key ecosystem service in mountains. The magnitude of this effect depends on avalanche intensity, forest attributes, and forest position within avalanche paths. Incorporating forest effects on avalanche runout into simulation tools is crucial, since quantifying these effects provides valuable data for risk assessments. While physical process-based simulation tools are predominantly used to model avalanche-forest interaction, empirical approaches are rare. However, empirical models are a good alternative in data-scarce areas and facilitate regional-scale analyses due to their computational efficiency.

The open-source avalanche simulation framework AvaFrame provides the empirical simulation module com4FlowPy for gravitational mass flows (GMFs). It routes a GMF along an identified path in three-dimensional terrain. GMF runout and intensity are determined by geometrical runout angle (α) concepts. The Forest Friction Module in com4FlowPy calculates the effect of forests on GMF transit and runout. Required input data include a Digital Elevation Model (DEM), GMF starting zone(s) and a forest layer. In forested areas, a GMF’s energy dissipation due to forest friction is modelled by increasing pre-defined α.

To offer a comprehensive understanding of Forest Friction parameters, we performed a sensitivity analysis. We analysed modelled avalanche runout distances from varied parameter settings under different initial and boundary conditions. Parameter variation was performed in scenarios with generic topographies of different steepness but with same forest extent, and avalanche intensities. The results show that Forest Friction parameter sensitivities depend on pre-defined α and terrain steepness. Greater parameter sensitivities were observed in flatter avalanche terrain (30°) compared to steeper terrain (45°). Modelled runout lengths reacted more sensitively to lower modelled intensities (< 66 m/s) than to higher intensities (66–72 m/s) at which an avalanche enters forest. The Forest Structure Index (FSI), which describes a forest’s protective characteristics (0=no protection, 1=optimal protection), had the greatest influence on avalanche runout compared to other parameters. We summarised these insights in recommended parametrisations for different terrain and avalanche intensities and applied them in a case study where observed avalanche events were back-calculated.