Abstract

Recent developments in hydrometeorological modeling aim toward a more sophisticated treatment of terrestrial hydrologic processes. The objective of this study is to investigate the role of lateral terrestrial water flows on the regional atmospheric and terrestrial water cycle in a humid region of... Show moreRecent developments in hydrometeorological modeling aim toward a more sophisticated treatment of terrestrial hydrologic processes. The objective of this study is to investigate the role of lateral terrestrial water flows on the regional atmospheric and terrestrial water cycle in a humid region of Central Europe. This is accomplished by evaluating model results from both the standard Weather Research and Forecasting (WRF) model and the hydrologically enhanced version WRF-Hydro, which allows for a more comprehensive process description of the interdependencies between water and energy fluxes at the land-atmosphere interface. To account for internal model variability, an ensemble for each model variant is generated for a 3-month study period in the summer of 2005. The regional water cycle in the simulation domain is investigated by evaluating the ensemble results with a joint atmospheric-terrestrial water budget analysis. We focus on six differently sized river catchments located in Southern Bavaria (Germany) and the southerly adjacent Eastern Alps, where simulation results are compared to observations of precipitation, near-surface temperatures, and streamflow. Most prominently, WRF-Hydro increases (near-) surface runoff and decreases percolation, resulting in a reduced total runoff amount. Accordingly, soil moisture storage and evapotranspiration are increased. Domain-averaged precipitation differences between WRF and WRF-Hydro are essentially related to differences in atmospheric moisture inflow and outflow, which is the signature of internal model variability and is potentially enhanced in the WRF-Hydro ensemble. Driven only at the boundaries of the outermost domain, the fully coupled model system shows good performance in the reproduction of observed streamflow. Show less