GRID-SIZE EFFECT ON WATERSHED HYDROLOGIC SIMULATIONS
Z. Yu

Interest in modeling large scale hydrologic systems in a large-scale watershed has prompted the scale problem of grid size - what is an appropriate grid size for the hydrologic simulation. A distributed watershed model, Basin-Scale Hydrologic Model (BSHM), with a series of DEMs (grid spacing from 120 to 3600 feet) was used to examine the effect of DEM grid size on the land surface representation and hydrologic simulation. Frequency distribution of slope tan(b) was calculated for each DEM grid size. Developed algorithm with one-direction flow in BSHM was used for the calculation of spatial distribution of runoff travel time and the prediction of simulated hydrographs for each grid size. The effect of grid size on the ground-water flow simulation was also examined by simulating water-level configuration and stream ground-water interaction. Overall, a 1200 feet grid size provides basic estimations in both surface-water and ground-water simulations. A 600 feet grid size was suggested to be an appropriate one in terms of the quality of simulated outputs and the amount of required computing time.

Journal of Hydrology & Technology, vol. 13, no. 1-4, p. 75-85, 1997.

Simulating the Basin Response to Single-Storm Events of Various Resolutions Described by a Mesoscale Meteorological Model
Zhongbo Yu

An approach of linking the Penn State-NCAR Mesoscale Meteorological Model (MM5) to the Hydrologic Model System (HMS) was implemented to simulate the river-basin response to single-storm events. Three MM5-domain resolutions, no nesting (36 km), double nesting (36-12 km), and triple nesting (36-12-4 km), were used for two single-storm simulations in April 1986 and May 1988. The storm patterns were well captured in the storm simulation with MM5. The analysis of the spatial variability of simulated precipitation was conducted in various resolutions for the Susquehanna River Basin and its sub-basins. The simulated precipitation of various resolutions was used to drive the hydrologic simulation of surface runoff. The results indicate that the storm simulation with triple nesting is better than these with no nesting and double nesting. Both hourly observed and MM5-simulated precipitations were used to drive HMS, a distributed hydrologic model system that includes soil hydrologic, surface-water, ground-water, and channel ground-water interaction models, for simulating the streamflow at the basin outlet in a sub-basin of the Susquehanna River Basin, the Upper West Branch. The streamflow simulation with observed precipitation was first calibrated to the observed streamflow. The simulated streamflow with the observed precipitation was compared to the simulated streamflow with the simulated precipitation of triple nesting. The discrepancies in modeling the streamflow response with the simulated precipitation of various resolutions were evaluated for the Upper West Branch.

Journal of Geophysical Research, vol. 104, 19,675-19,690, 1999.