Effects of Climate Change on the Probability of Urban Tree Failures from Wind Gusts
Abstract
Trees grown in urban environments provide environmental, economic and psychological benefits to their surrounding communities. However, urban trees also pose significant risks since damaged trees can cause serious harm to people, housing, and infrastructure by falling on sidewalks, roads, houses or power lines. To better understand the risk posed to trees by wind, models have been developed that estimates the required wind speed needed to damage a tree or group of trees, and the likelihood that such a wind speed is met or exceeded annually. The importance of such models is rising each year as the associated risk grows as well, due to an increase in urbanization, frequency and intensity of wind storms increasing with global warming and growing evidence that elevated atmospheric CO2 concentrations, driven by climate change, cause trees to grow faster and larger, likely increasing their fragility to wind. In this thesis, a model was created to consider the impacts of climate change on trees’ risk using analysis of wind trends globally and locally in the Toronto region, and by considering the impact of the steadily increasing concentration of CO2 in the atmosphere. The CO2 increase impact on trees has been inferred based on meta-analysis data from 219 papers studying the impact of elevated CO2 growing conditions on 293 tree samples of varying age and species. The model functions by estimating the return period of wind storms that can damage an individual tree via trunk rupture or overturning. Meta-analysis data indicates that the density of leaves in tree crowns is likely to change with elevated CO2 concentrations. The aerodynamic impact of this change is currently not well understood. In an effort to improve the model further, experimental wind testing was conducted at the Wind Engineering, Energy and Environment Dome (WindEEE) at Western University, where a 9-year-old, 1.9 m tall red maple (Acer rubrum) was subjected to wind speeds from 6-12 m/s. The testing was repeated 5 times, between each repetition the crown was thinned by 25% to simulate varying crown leaf densities of the tree, and to analyze the relationship between the density of leaves in the crown and the drag coefficient.