<div> <p><span>Large wood (10cm diameter & 1m long) gets recruited into a mountain river system from surrounding forested areas. Instream large wood positively influences the diversity of the river system, creating habitats for terrestrial and aquatic species. However, the corresponding risk to the presence of instream large wood is a more controversial topic in river management. On the one hand, large wood increases the riverbed roughness, partly dissipating energy during a flood. On the other hand, its transport during floods might cause damage to infrastructure. Direct observations or monitoring stations are scarce and knowledge on how and when wood is transported remains far from complete.</span><span><br></span></p> </div><div> <p><span>In order to quantify a river’s instream wood transport regime, w</span><span>e are developing a video-based wood tracking system that counts the number of pieces that pass a certain point and estimates their sizes. We use a DeepSORT algorithm that uses machine learning to identify individual pieces of instream wood and draws a bounding box around every piece. Subsequently, it uses a Kalman filter to estimate the piece’s trajectory. To prevent counting the same pieces multiple times, the projected trajectory is compared to the detections in the subsequent frame. The system is designed so that it can be applied to different datasets and will be available to the increasing wood monitoring efforts around the world. For a more detailed look into the large wood regime at one of our main study sites (Vallon de Nant, Switzerland), and to calibrate our video-based wood tracking system, we have installed RFID tags into all pieces of large wood (approximately 1000 pieces) over a stretch of 3 km. A stationary RFID antenna registers the tagged pieces that pass by, of which the size and origin are known.</span><span> <br></span></p> </div><div> <p><span>First results show that the custom trained DeepSORT algorithm can not only identify pieces of instream wood, but also largely follow the pieces in subsequent frames. The approach seems to outperform current computer vision solutions. In our ongoing research, we aim to make the system more robust and expand the observation network to other rivers. With an expanding dataset, containing (manually) labelled training samples from different locations, and the low-cost measurement set-up, the system promises to aid successfully to an intercomparison of river systems in the context of the wood management debate.</span><span><br></span></p> </div><div> <p><span>This work is supported by the SNSF Eccellenza project PCEFP2-186963 and the University of Lausanne.</span></p> </div>

The colonization of proglacial margins by vegetation following glacier recession is a slow process, not least because glacially produced sediments are commonly well drained. Following from human‐induced climate change, warming could increase both growth rates and water availability because of glacier melting, so compensating for situations where climate change reduces precipitation. Compensation is likely a function of location, which will control access to meltwater and groundwater, themselves spatially variable. For the Olguin glacier (Torres del Paine, Chile), we test the hypothesis that as climate has warmed and precipitation has fallen, tree growth rate response is dependent upon the access of trees to glacial meltwater. Cores were taken from trees in three revegetating zones: (Z1) proglacial stream proximal, (Z3) proglacial stream distal, and (Z2) intermediate between Z1 and Z3. For trees within each zone, we measured annual tree‐ring widths and δ2H values. Z1 growth rates were strongly correlated with temperature and Z3 with precipitation, and Z2 showed a shift from precipitation correlation (i.e., following Z3) to temperature correlation (i.e., following Z1) through time. δ2H values were lowest at Z1, reflecting water of glacial origin, were highest at Z3, reflecting meteoric water supply, and shifted through time at Z2 from meteoric to glacial. Increased water supply associated with temperature‐driven glacier recession may compensate for decreasing water supply from precipitation to influence tree growth. This compensation is likely related to the spatial organization of the subsurface flux of glacial melt and leads to different revegetation processes to those envisaged in the classical chronosequence model of vegetation following glacier recession.

No abstract

<div> <div> <p><span data-contrast="auto">By creating pools and retaining sediment and organic matter, instream wood provides habitats for a vast variety of different species. It creates a complex river bed and is essential for a healthy ecosystem (Wohl et al., 2019). However, during extreme weather conditions, floods can mobilize the wood and transport it, causing a hazard to downstream infrastructure. Therefore it is important better understand river wood dynamics, such as storage and transport regimes. These regimes are influences by individual log characteristics (e.g. shape, density and orientation), but also individual river weather, climate and geographical factors. In the last decade, an increasing amount of case studies have been performed, although still limited in amount of logs tracked in European rivers (Wyzga et al., 2017). In our current contribution, we deploy a tracking and monitoring system in an Alpine river in the canton of Vaud, Switzerland. The Avancon the Nant is located in the Vallon de Nant, a valley that has been protected since 1969 (Vittoz and Gmür, 2009), and can therefore be argued to have a close to natural wood regime.</span></p> <p><img src="" alt="" /></p> <div> <div> <p><span data-contrast="auto">Figure: Locations of instream wood in 2022 as compared to 2021. In grey, 3 special sections (wider sections and sections with multiple streams) of river are represented.</span></p> </div> </div> <div> <div> <p><span data-contrast="auto">In the summer of 2021, 948 (0001 to 0948) pieces of instream wood were tagged with a unique number and 2 unique RFID tags. One year later, in another field campaign, the movement of the pieces was assessed (see figure). From the pieces that have been recovered (7% were lost), a total of 20 pieces were found to have moved with an average of 260 meters. These movements took place in specific sections, primarily in single-threaded narrow sections. The two lower special river sections (w1 and w2) were found to contain pieces with a larger diameters as compared to the other sections. As the tree density decreases when moving up the river, also the total volume of wood storage and the amount of pieces decreased. Furthermore, more pieces with a high degree of decat were found as compared to fresher pieces. This indicated that in recent years, less wood recruitment has taken place.</span><span data-ccp-props="{"> </span></p> </div> <div> <p><span data-contrast="auto">REFERENCES </span><span data-ccp-props="{"> </span></p> </div> <div> <p><span data-contrast="auto">Vittoz, P., & Gmür, P. 2009: Introduction aux Journées de la biodiversité dans le Vallon de Nant (Bex, Alpes vaudoises), </span><span data-contrast="auto">Mémoire de la Société vaudoise des Sciences naturelles, 23, 3-20.</span><span data-ccp-props="{"> </span></p> </div> <div> <p><span data-contrast="auto">Wohl, E., Kramer, N., Ruiz-Villanueva, V., Scott, D. N., Comiti, F., Gurnell, A. M., Piegay, H., Lininger, K. B., Jaeger, K. L., Walters, D. M., & Fausch, K. D. 2019: The natural wood regime in rivers, BioScience, 69, 259–273.</span><span data-ccp-props="{"> </span></p> </div> <div> <p><span data-contrast="auto">Wyzga, B., Mikus, P., Zawiejska, J., Ruiz-Villanueva, V., Kaczka, R. J. & Czech, W. 2017: Log transport and deposition in incised, channelized, and multithread reaches of a wide mountain river: Tracking experiment during a 20-year flood, Geomorphology, 279, 98-111.</span><span data-ccp-props="{"> </span></p> </div> </div> </div> </div>

<p>Bern University of Applied Sciences, School of Agricultural, Forest and Food Sciences, COSCI, Hydraulic Platform LCH, Institute of Civil Engineering, EPFL-ENAC, Lausanne, CH and University of Lausanne, Institute of Earth Surface Dynamics (marceline.vuaridel@unil.ch)</p><p>Floods and intense surface runoff are recurring hazards known for triggering erosion processes at the channel and the catchment slope scales, respectively. Whilst the firsts determine the removal of streambank material, also referred to as hydraulic streambank erosion (e.g., Ruiz-Villanueva et al., 2014), the seconds are typically responsible for destabilizing shallow landslides. Both processes are exacerbated by extreme precipitation events, and can cause important damages to forests, agriculture, civil structures, and settlements through the loss of land masses. Moreover, streambank erosion and shallow landslides can be responsible for the recruitment of large wood (LW), whose transport during floods may strongly impacts on downstream infrastructures of urbanized areas (e.g Ruiz-Villanueva et al., 2014).</p><p>Via augmented mechanical stabilization, plant roots may significantly decrease the susceptibility of riverbanks to hydraulic erosion as well as shallow landslides. Under certain conditions, plant roots can be considered an alternative protection against such processes with respect to other civil engineering measures (Stokes et al., 2014). However, root reinforcement effectiveness depends on many factors such as roots density, soil properties, and soil thickness (Cohen and Schwarz, 2017), which implies that some vegetated areas have a more significant effect than others. Most available models ignore the contribution of plant roots with acceptable spatial resolution.</p><p>In this work, we present BankforNET and SlideforNET, two physically-based modelling tools, which have been developed to take the different stabilizing effects of soil reinforcement mechanism by plant roots into account. This is important for proper modeling of bank erosion and landslides processes during extreme events, and to optimize forest protection strategies. BankforNET is a one-dimensional, probabilistic model which simulates expected hydraulic streambank erosion by considering channel morphology, bank sediment material, vegetation roots, and a certain discharge scenario. The SlideforNET is a probabilistic model based on the 3D analysis of slope stability and takes the lateral and basal root reinforcement into account. Ultimately, it gives an estimation of the degree of protection of a forest against landslides.</p><p>These tools are currently being tested in a catchment of 29 km<sup>2</sup> in NW Switzerland for the priorisation of protective forests against risks related to LW transport during floods. Based on the model results, the possible silvicultural measures are defined considering quantitative criteria such as the risk mitigation effect of the forest stands, or their risk increment due to LW recruitment and transport. This study is an example of how quantitative tools can be use by decision makers to priories the role of protection forest in a catchment and to support the definition of silvicultural measure to mitigate the risks due to LW transport.</p>