Spatially-distributed snowmelt, water balance and streamflow modelling for a large mountainous catchment: Boise River, Idaho, USA

As a demonstration of the potential of spatially-distributed hydrological modelling for large catchments in mountainous areas, a snow simulation model and a water balance/streamflow model have been linked together and applied to a 2150 km(2) portion of the Boise River catchment in Idaho, USA. The snow model simulates all of the energy fluxes into and out of the snowpack and requires spatial field time series of all of the major meteorological inputs. It is being run at 250 in grid resolution with a 3-h time step. The output spatial fields of this model are aggregated in space and time and are used as input to the water balance model. The water balance model simulates all of the major land surface processes using algorithms that are physically based but are constructed so as to be appropriate for this large spatial scale. It makes extensive use of geographic information systems for parameterization. The model operates on a daily time step and can have a variable cell size; for the Boise River, a I km(2) grid cell size is being used. These models are intended to address water and natural resource management issues, but the concepts of finding appropriate mathematical formulations and of matching the modelling scale with the process scale have broader significance.