The present thesis begins with an overview and different classifications of mass movements. Debris flows, one type of mass movements, are described in detail. First, their characteristics are presented and also classified according to the type of triggering, magnitude and velocity. The main differences between the debris flows and other mass movements are also described. In addition, the mathematical model Flo-2D is described. Its use, functioning and procedure of modeling debris flows are presented, as are the previous experiences with debris flow modeling in Slovenia and the use of different mathematical models, including Flo-2D. In the following chapter, the Slovene legislation in the area of flooding and erosion hazard is described. This legislation prescribes the methodology of preparing the preliminary hazard maps, hazard maps and risk maps due to the influence of flooding and erosion. The thesis proposes a protocol of preparing the preliminary hazard maps and hazard maps due to the debris flow effects. Debris flows are divided into torrential and slope debris flows. The main part of the thesis is dedicated to the practical use of the mathematical model Flo-2D in debris flow modeling of torrential fans using only the publicly available data. The alpine valley of Sava Dolinka River was selected as a test region. In the sensitivity analysis of the Flo-2D model, the influence of different parameters was tested: the control parameters settings of the model, the position of the incoming hydrograph, the Manning roughness coefficient, the computational grid size, different rheological characteristics and the influence of different ways of modeling objects on torrential fans. The possibility of using the results of the models made with publicly available data is evaluated in preparing hazard maps. This method would enable the preparation of hazard maps with relatively low costs and without the expensive extra surveys. The possibility of preparing hazard maps using an application proposed in the Flo-2D model is also presented. The topographic data DEM5 turned out to be the main deficiency of all the input data used in the research, as they can dramatically affect the model results and hazard area assessments. The rest of the applied model input data can be defined well enough to be efficient in preparing the hazard maps the way it is described in this thesis.