Tree acclimation to wind in a dense beech forest to investigate how mechanical cues are filtered and modify radial growth.

Climate change stresses the need of a better understanding of the environmental control of forest growth. Wind is an important climate factor well known to influence plant growth. More precisely, experiments where plants were bent in controlled conditions have proved trees can sense wind-induced strains and adapt their growth to, which is known as thigmomorphogenesis (Moulia et al., 2011). However these types of experimental conditions are often too simplistic compared to the complex pattern of mechanical stimulus experimented by trees in natural conditions. In greenhouse experiments, the mechanically stimulated plants are usually submitted to large and frequent amount of mechanical strain and compared to fully protected plants in order to emphasize thigmomorphogenetic responses. The very few field experiments consisted in guying trees and comparing their growth with control trees after several months. These studies have proved that mechanical stimulation of growth allocation is an important ecological process that can impact strongly forest ecosystem services as wood production, carbon storage or landscape protection against storms and strong winds. Actually , thigmomorphogenesis could account for 50% of the aerial growth (Meng et al., 2006). However the stimuli – i.e. the wind-induced strains in living cells - were not quantified and the time resolution (one to several years) was too low to give more insight about how thigmomorphogenesis takes place. This work aims at accurately quantify thigmomorphogenetic responses of radial growth in a broaleaved forest (Fagus sylvatica, Northern Eastern France, public forest managed by ONF). We will particularly focus on wind perception by the trees and possible filtering of the different type of wind events by trees. We also want to study the between-tree variability of the mechano-sensing and growth response in relation with social status (dominant and suppressed trees) in very dense stands, undisturbed for a long time (no thinning). Assuming trees in this undisturbed stand are acclimated to usual winds, we monitored wind-induced strains in several trees of different status, then we manipulated this mechanical signal by adding or suppressing a giving amount of this natural stimulus (by guying or pulling some trees) and observed radial growth response