A dendroprovenance approach to fingerprint the origin of instream wood at the river basin scale

Instream large wood (LW) plays an important role in the geomorphic and ecological diversity of a fluvial ecosystem. However, during flood events, LW can also pose a risk to infrastructure and populations by blocking channels, damaging bridges, and other structures. The primary source of LW are forested areas established along rivers and their upstream network. Understanding the origin of LW and the factors that influence its dynamics is key for optimizing river and riparian forest management and reducing the risk associated with flood events.

We study a 50km reach of the Rhone River between the city of Geneva and Génissiat dam (France), where the wood material arriving is retained. In this 3000km² catchment, we aim to infer the origin of the LW that arrives at the reservoir by differentiating between the two main tributaries and providers of LW in the reach: The Arve River (coming from the Alps Mountains) and the Valserine River (from the Jura Mountains).

We have explored several methods for inferring the origin of instream wood. By combining them, we gain a more comprehensive understanding of the factors that influence the supply of LW to the watershed and its dynamics within the river system. These methods are based on:

• Stable isotopes in the cellulose coming from the water molecule (δD and δ¹⁸O): they present a spatial distribution due to fractionation happening during evaporation-precipitation processes. The tree absorbs the isotopic signal and stores it in the cellulose, that can be analyzed to distinguish between different source areas.
• Chemical composition of wood cellulose: it can be analyzed to provide information about the geology of the area where the tree grew. Techniques such as inductively coupled plasma optical emission spectrometry (ICP-OES) or X-ray fluorescence spectrometry (XRF) are used for this purpose.
• Riparian forest composition: some forest characteristics (e.g., tree species, tree diameter, forest density, dead wood present in the floodplain, lateral connectivity with the river, etc.) provide useful information on the areas that are supplying the wood that reaches or will potentially reach the dam.

By combining these approaches, it may be possible to distinguish between different source areas of LW within the catchment and to better understand the factors that influence the supply of LW to the river system. We aim to develop a method that can be applied to similarly scaled mountainous catchments to determine the origin of instream large wood.