Quoted from: https://vic.readthedocs.io/en/master/ 

The VIC model (Liang et al., 1994) is a large-scale, semi-distributed hydrologic model. As such, it shares several basic features with the other land surface models (LSMs) that are commonly coupled to global circulation models (GCMs):

1. The land surface is modeled as a grid of large (>>1km), flat, uniform cells
   • Sub-grid heterogeneity (e.g. elevation, land cover) is handled via statistical distributions
2. Inputs are time series of sub-daily meteorological drivers (e.g. precipitation, air temperature, wind speed, radiation, etc.)
3. Land-atmosphere fluxes, and the water and energy balances at the land surface, are simulated at a daily or sub-daily time step
4. Water can only enter a grid cell via the atmosphere
   • Non-channel flow between grid cells is ignored
     • The portions of surface and subsurface runoff that reach the local channel network within a grid cell are assumed to be >> the portions that cross grid cell boundaries into neighboring cells
   • Once water reaches the channel network, it is assumed to stay in the channel (it cannot flow back into the soil)

This last point has several consequences for VIC model implementation:

1. Grid cells are simulated independently of each other, there is no communication between grid cells
2. Routing of stream flow is performed separately from the land surface simulation, using a separate model (typically the routing model of Lohmann et al., 1996 and 1998)