Culverts play a critical role in conveying surface runoff during flash flood events, yet their failure due to blockages can significantly alter flood dynamics, particularly in small, topographically complex catchments. Despite this, culvert blockages are often neglected in flood modeling. To address such gap, this study presents a comprehensive analysis of culvert blockages using the open-source hydrodynamic model TELEMAC-2D, applied to a flash-flood-prone catchment in central Germany. First, the study assesses recent flood events and evaluates the completeness and accuracy of official culvert datasets, identifying missing culverts. A dynamic culvert blockage module is then implemented, simulating varying degrees and timings of blockage based on water level thresholds at culvert inlets. Through a series of flood scenarios, the study identifies culverts whose blockage has a major impact on flood hydrographs and inundation extents. Results highlight the importance of accurate culvert representation and demonstrate how scenario-based modeling of blockages can support the identification of critical infrastructure. This enables the development of targeted mitigation strategies, such as prioritized maintenance or emergency protection, ultimately reducing flood risks. The findings underscore the need to integrate culvert blockages into flash flood modelling and risk assessments and support future research into blockage formation mechanisms and improved field data acquisition.