Résumé
A better knowledge on the diversity of wood would have implications in several domains of pure and applied research. Large databases of wood physical-mechanical properties are the pre-requisite for applying material selection method, which could allow more efficient, diversified and potentially sustainable utilisation of wood materials. They can also provide insights into the physical understanding of traditional preferences for specific woods in various cultural uses. In addition, wood diversity is increasingly taken into account by research into biomechanics and functional ecology, as properties and their microstructural / chemical affecting factors can reflect growth strategies related to taxonomy and/or biogeography. Exploring the diversity of wood properties requires obtaining very large datasets. However, if a significant number or species have been characterised for some properties (such as density or modulus of elasticity along the grain), information is still scarce on other important aspects of wood behaviour (such as viscoelasticity and anisotropy). The present work aims at contributing to overcome this lack of knowledge. Experimental characterisations of dynamic modulus of elasticity and of damping coefficient of many lesser-known wood species are combined with an extensive compilation of data scattered in the literature. This new "viscoelastic vibrational properties of wood" database, which covers 450 species, is linked to the CIRAD database of technological properties of tropical woods which covers 1000 species. The global dataset covers 1310 species. Observed correlations between properties are discussed in connexion to affecting factors such as porosity, orientation of wood elements, presence of extractives. Properties distributions are compared between softwoods, temperate or tropical hardwoods, and between botanical families. The new dataset on vibrational properties could be used to select appropriate species for applications requiring specific ranges of damping. (Résumé d'auteur)