Background and Aims <p>Root structural defects are common in urban trees due to natural reasons or human activities. However, the influence of such defects on the trees’ capacity to resist uprooting failure is seldom considered, because the numbers and positions of defects within underground root systems are usually unknown and uncertain. To consider this uncertainty, a stochastic Material Point Method (SMPM) is developed for simulating large-deformation tree uprooting involving potential root structural defects.</p> Methods <p>In the proposed SMPM approach, a Poisson process is used to stochastically simulate the occurrence of structural defects within root systems, and an advanced Material Point Method (MPM) model is adopted to simulate large-deformation uprooting behavior. Under the framework of Monte Carlo simulation, SMPM simulations of tree uprooting are performed repeatedly to reveal how the uncertainty in root structural defects is propagated and affects the tree uprooting resistance.</p> Results <p>A probability distribution of tree uprooting resistance is obtained to quantify tree failure risk and facilitate decision-making. Assessment of tree stability without considering root structural defects and their uncertainty could be misleading. The simulation results suggest that defect location, particularly proximity to the root base, may exert a stronger control on uprooting resistance than defect number alone.</p> Conclusion <p>For the first time, a stochastic approach is introduced to tree stability analyses for modeling uncertainties induced by potential root structural defects. To reduce the probability of uprooting failure, cutting or damaging roots near the trunk should be avoided, and proper drainage around the root collar should be ensured.</p>