Anotace:
Flooding is a natural phenomenon that occurs with varying intensity and at irregular time intervals. Floods are the natural disasters that pose the greatest direct threat to the Czech Republic. They may cause serious critical situations during which not only extensive material damages are incurred, but so too is the loss of human life in affected areas as well as vast devastation of the cultural landscape including environmental damages. The information issued by flood forecasting services about the character and size of flood areas for individual N-year flood discharge events and specific flood scenarios is important for eliminating the potential threats and consequences of such events. Hydrodynamic models provide an adequate image of depths and flow velocities at the longitudinal or cross profiles of the watercourse during a flood event. This is why information obtained from hydrodynamic models occupies a privileged position from the viewpoint of protecting human life and mitigating property damage. Altimetry data are the basic input into hydrodynamic models. One way to obtain such data is through the method of aerial laser scanning (ALS) from the digital terrain model (DTM). This method is considered one of the most accurate methods for obtaining altimetry data. Its major drawback is however its inability to record terrain geometry under water surfaces due to the fact that the laser beam is absorbed by the body of water. The absence of geometric data on watercourse cross sectional area may perceptibly affect results of modelling, especially if the capacity of a missing part of the channel represents a significant cross sectional area. One of the methods for eliminating this deficiency is sufficiently calculating channel depth by means of software tools such as CroSolver. This paper deals with the construction of a hydrodynamic model using fifth generation DTM data and compares outputs from this model at various discharges with a model based on the altimetry data modified using CroSolver. Outputs from the two hydrodynamic models are compared using HEC-RAS software with the use of depth estimate data and with the use of the unmodified DTM. The comparison is done on two watercourse reaches with different terrain morphology and watercourse size. A complementary output is the comparison of inundation areas issuing from both model variants. Our results indicate that differences in the outputs are significant, namely at lower discharges (Q1, Q5), whereas at Q50 and Q100 the difference is negligible with a great role played by the morphology of the modelled area and by the watercourse size.