SmartWood: Laboratory experiments for assessing the effectiveness of smart sensors for monitoring large wood movement behaviour

Abstract

Large wood (LW), delivered to the river channel in the course of commercial forest harvesting, or generated during natural events, can be mobilised during floods. The movement of wood along the river corridor involves complex cycles of recruitment, mobilisation, transportation and deposition. These processes are affected by the size, buoyancy, roughness and complexity of the wood components, as well as the relative spatial density and the character of the channel boundary elements. In order to understand the probabilistic behaviour of woody elements within the fluvial system, it is important to be able to characterise the timing, mechanisms and duration of the various phases of wood transport. Due to a lack of suitable sensing technology, a detailed understanding of LW recruitment, transport and accumulation processes has thus far been elusive. In this study we introduce a technique using a nine-degrees of freedom (9-DoF) sensor embedded in a ‘SmartWood’ dowel that shows strong potential for measuring and recording LW movement. The SmartWood assembly comprises an integrated sensor with an inertial measurement unit (IMU), accelerometer, gyroscope and magnetometer, installed in a wooden dowel that is scaled to represent a tree stem in the flume. The sensor is able to record the many different motions of LW transport. The sensor-tagged wood dowel is density-compensated, with a specific weight of 0.5 g·cm−3. A series of verification and experimental tests was carried out to evaluate the applicability of this new technology for LW research and is presented herewith. Experiments were conducted in a 6.3 m long and 1.5 m wide flume with sinuous channel course and mobile gravel bed conditions, with a discharge of up to 10 l·s−1. We show that LW movement during transport, particularly starting, rolling, yawing and stopping processes, but also LW impacts, can be quantified within a flume environment. These findings can be further developed to obtain the translational movement behaviour of LW, which is needed to refine probabilistic models of downstream trajectories. Understanding complex LW movement is essential for informing freshwater and forestry management guidelines.