This paper focuses on the impact of geotextile tube damage at the joints between tubes on dam structures subjected to seepage. First, a seepage-induced failure critical gradient model for damaged geotextile tubes was developed. Tests were conducted using geotextile specimens with precut O-shaped holes to simulate the seepage erosion process. Various overburden loads (0, 5, 10, 20, and 30 kPa) and hole radii (0.5, 1.0, 1.5, and 2 cm) were examined. Based on the test phenomena and the changes in pore-water pressure and seepage flow, four progression stages (seepage stability, sand particle wash-out, preferential flow formation and development, and complete failure) were identified. The experimental critical gradients obtained under different conditions agreed well with the model results. The critical gradient is positively correlated with the overburden load and negatively correlated with the hole radius. Critical gradient growth gradually slows with increasing overburden load. The critical gradient difference caused by the hole size decreases rapidly. When the overburden load increases to 20 kPa, this difference is essentially unchanged. These findings can provide a better understanding of the performance of damaged geotextile tubes.

Urban flooding has become one of the most common natural hazards threatening people’s lives and assets globally due to climate change and rapid urbanization. Hydraulic structures, e.g., sluicegates and pumping stations, can directly influence flooding processes and should be represented in flood modeling and risk assessment. This study aims to present a robust numerical model by incorporating a hydraulic structure simulation module to accurately predict the highly transient flood hydrodynamics interrupted by the operation of hydraulic structures to support object-level risk assessment. Source-term and flux-term coupling approaches are applied and implemented to represent different types of hydraulic structures in the model. For hydraulic structures such as a sluicegate, the flux-term coupling approach may lead to more accurate results, as indicated by the calculated values of NSE and RMSE for different test cases. The model is further applied to predict different design flood scenarios with rainfall inputs created using Intensity-Duration-Frequency relationships, Chicago Design Storm, and surveyed data. The simulation results are combined with established vehicle instability formulas and depth-damage curves to assess the flood impact on individual objects in an urbanized case study area in Zhejiang Province, China.