The main objective of this paper was to investigate how the errors of the Digital Elevation Model (DEM) can affect the results of the hydraulic modeling. We investigated the impact of DEM on the flood areas and flood depths, which from the point of view of the stakeholders who make decisions concerning the flood protection are crucial. The second objective of this paper was evaluation of methods of limiting the flood risk assessment uncertainty caused by the DEM errors. DEM developed within the ISOK project and DEM developed within the LPIS project were studied. In order to meet the objectives, we conducted analyses of the DEM errors in the test area of the Warta river section. In the next step, hydraulic, one-dimensional models were developed in DHI Mike 11 software, based on the two DEMs. Based on these hydraulic models results (water levels), the floodplains were delineated. Reaches and water depths of the floodplains were analyzed in order to find out, how the errors of the studied DEMs affect assessed flood risk. In the case of the less accurate LPIS DEM, two methods of managing the DEM error- based uncertainty were adopted: the method assuming an uniform distribution of DEM error and the Monte Carlo method. The results show that the error of the LPIS DEM (which was about 0,7 meter in the study area) may lead to the underestimation of the flood risk and thus to a wrong decision concerning the location of the flood protection actions. The method assuming an uniform distribution of error does not limit this hazard enough, while the Monte Carlo method allows to manage the flood risk due to the construction of the flood probability map.