2503 River Flow 2014 – Schleiss et al. (Eds) © 2014 Taylor & Francis Group, London, ISBN 978-1-138-02674-2 Wood density assessment to improve understanding of large wood buoyancy in rivers Virginia Ruiz-Villanueva & Markus Stoffel Dendrolab.ch, Institute of Geological Sciences, University of Bern, Switzerland Institute for Environmental Sciences, University of Geneva, Switzerland Hervé Piégay, Vincent Gaertner & Franck Perret National Center for Scientific Research (CNRS), UMR 5600 EVS/Site ENS de Lyon, France ABSTRACT: Drift wood (or in-stream large wood, LW) plays an important role in river ecosystems by influencing hydrodynamics and morphology. Wood density, therefore, buoyancy, is the main factor conditioning initial motion of wood in rivers. The better understanding of wood density and the dry-wet process and decay, may be useful to improve the knowledge of wood dynamics in rivers. We analyze wood pieces retained in a dam, in the French Rhone, and a set of freshly cut riparian trees. Different protocols were set to measure density and buoyancy of these two series of wood and to test effects of drying and wetting, species and wood decay stages. Preliminary results after seven months of experiments show dif- ferent behaviors in density and buoyancy depending on species. Light wood (360–500 kg ฀ m 3 ) is likely to show a buoyancy rate about 52% (half of the log emerged), medium density wood (500–700 kg  m 3 ) about 39%, 21% for dense wood (700–900 kg ฀ m 3 ) and 12% for very dense wood (900 kg  m 3 ). We observed a significant negative correlation between wood density and buoyancy and proposed a model to predict wood buoyancy. The results from this work will help to understand the evolution of buoyancy through time and estimate local conditions of entrainment and transport of wood in rivers. varies greatly among tree species, but it also varies for each of the species according to water content and degree of decay. Therefore, the time during which wood pieces are wetting or drying can affect their dynamics when they enter the river. The final goal of this work is to improve our understanding of wood buoyancy in rivers throughout the assessment of wood density for different types of wood and decay stages and to find a model to predict wood buoyancy. 2 STUDY SITE AND METHODOLOGY 2.1 Sampling The characteristics of the wood are assessed using two series of wood pieces, one extracted from a reservoir (decayed floating wood) and another from living trees used as a reference (green or fresh wood). The decayed wood analyzed was collected from wood pieces retained in the Génissiat reser- voir, French Rhone (watershed area of 10,910 km 2 at Génissiat). This gravity dam has no overflow pathway so that all wood coming from two main tributaries, the Arve and Valserine Rivers, is blocked by the dam, even during floods, and must be extracted mechanically (Fig. 1). 1 INTRODUCTION In-stream Large Wood (LW) plays an important role in river ecosystems by influencing hydrology, hydraulics, sedimentology, and morphology (Montgomery, 2003). An extensive literature now exists describing the influence of wood on stream ecology (Gregory et al., 2003; Kasprak et al., 2011), and more recently on stream geomorphology (Gurnell, 2012; Wohl, 2013). Recent research has focused on the mobilization of woody material during floods (Comiti et al., 2012), as transported woody material can cause a substantial increase in the destructive power of floods (Ruiz-Villanueva et al., 2012). Various characteristics of a piece of wood affect its likelihood of movement (e.g., wood density, buoyancy, orientation, size, and form related to flow depth, velocity, and roughness; Le Lay et al., 2013). In this study, we analyze wood buoyancy to determine whether floating or sinking of wood pieces through the water column has a funda- mental impact on river dynamics. Buoyancy typi- cally varies with tree species and decay rates. So, the most important property affecting river wood mobility and particularly the capacity of LW to float in freshwater is wood density. Wood density