Model Construction:Describing overland flow
The computational method used to compute overland flow is selected in the GSSHA Job Control Parameters dialog from Overland flow sim. Three methods are available.
- Explicit – alternating direction, time varying version of the original point explicit method developed for CASC2D, as described by Julien and Saghafian (1991).
- ADE – alternating direction explicit (Downer 2002).
- ADE-PC – alternating direction explicit with prediction-correction (Downer 2002).
The default value is Explicit. The ADE and ADE-PC methods are described in the GSSHA User’s Manual. The explicit method has a variable time step that can adapt to computational needs. The ADE-PC method is very robust and may be employed when particularly difficult conditions are encountered. The ADE-PC method will often work when the other two methods will not. The additional computations in the ADE-PC method make it significantly slower than the other two methods, which require about the same wall clock time. Some experimentation may be required to determine which method will work best for a particular problem.
The following inputs are required in overland flow simulations in GSSHA.
- Land surface elevation.
- Land surface roughness.
The grid cell land surface elevations (determined from the DEM, as discussed in Chapter 2) and the surface roughness comprise the minimum input parameters that must be defined for a GSSHA surface runoff simulation.
The surface roughness represents the overland Manning’s roughness coefficient n. These values can be spatially distributed using an index map defined from vegetation cover and/or land use. Values of overland roughness coefficients based on vegetation coverage are presented by Engman (1986) and Ree, Wimberly, and Crow (1977), and summarized in the GSSHA User’s Manual (Downer and Odgen in preparation).
By using Manning’s resistance equation, it is assumed that the overland flow is turbulent flow over rough surfaces. Manning’s roughness coefficients are dimensionless. Assignment of parameter values to every grid cell is discussed in Chapter 5.
- Model Construction