Where gauged rainfall/runoff data is available for a range of events it should be used in preference to the above regression equation with modifying factors.
The original regression equation (Equation (8)) does not differentiate between catchments with the same degree of urbanisation but different roughnesses. An additional empirical parameter has therefore been added to take pervious sub-catchment roughness into account.
The parameter PERN is inputted as a Mannings 'n' representation of the average sub-catchment roughness. B is then modified in accordance with the following table. If PERN is left blank then B is unchanged.
Note: In urban catchments it is recommended that individual sub-catchments be split into separate impervious and pervious components. This is achieved by utilising both the first and second area dialogs. The first can relate to either the impervious or pervious, however you need to be consistent to arrive at proper component totals in the output. If OSD is being applied it is necessary to associate the first area dialog with the impervious component of the sub-catchment.
It is common for a split sub-catchment analysis to estimate a lower sub-catchment peak than a peak using only a lumped (impervious plus pervious component) sub-catchment definition.
Based on a study calibrating urban catchments in Canberra the surface runoff routing parameters for PERN Manning's roughness for impervious and pervious areas were 0.015 and 0.040 respectively. (Willing and Partners, 1993)
During calibration of a gauged catchment an additional parameter BX in the header data is included to modify the calculated or input B by a further multiplication factor. The parameter BX will then uniformly modify all sub-catchment B values previously computed, or set (Coefficients B and n).
Note: The BX value by default is set to 1.0. This value (the Storage Coefficient Multiplication Factor in the Job Control dialog) can be varied to provide a lumped calibration for the outflow peak at the catchment outlet.