How does uncertainty in digital elevation data obtained by laser scanning affect the outcomes of hydrological modelling?

The digital elevation of Sockholm city has been obtained by laser scanning with the density of measurements every 1 meter. These topographic data contain heights and cover all surface, except for buildings and some forest stands or trees. The data have a measurement error of 5-12 cm. While the spatial resolution of the data is very high, quantitative calculations for hydrological modelling, such as flow direction and accumulation, as well as identification of watersheds, are influenced by the error inherent in the measurements. To be able to model drainage networks, first a continuous digital elevation model has to be calculated using interpolated heights to fill the locations that lack measurement data. The influence of varying uncertainty in elevation data to calculation of flow and watersheds can be estimated using Monte Carlo simulation. The aim of the project is to use GIS to develop techniques for visualization and presentation of uncertainty in raster digital elevation data. Both randomly distributed and known estimation errors in a digital elevation model are to be modelled and mapped.
Software: ArcGIS, Idrisi32.
Language: English or Swedish.