|
|
|
|
|
|
|
token, the capacity of the main drain of the study area was itself exceeded during floods, and substantial amounts of runoff would then cross the area boundary into a neighbouring catchment. This phenomenon was clearly observed on several occasions, and was incorporated in the model. |
|
|
|
|
|
|
|
|
14.2.2
Selection of Hydraulic Model |
|
|
|
|
|
|
|
|
The dynamic drainage simulation model SPIDA (developed by Wallingford Software) was chosen to model the Pardeshipura catchment. SPIDA was chosen because it has some capacity to model flooding, and it has a suitably simple way to incorporate the effects of solids deposition. In addition, SPIDA is a mature hydraulic model that has been widely used and tested, reducing the risks of programming errors inherent in building new software. |
|
|
|
|
|
|
|
|
With SPIDA, the user specifies the depth of solids at each section of the channel. The model then treats this depth as a fixed physical obstruction, with a user-specified sediment roughness. Hydraulic computations thus reflect both the reduced hydraulic cross-section and the increased hydraulic roughness resulting from the deposition of solids in each section. SPIDA assumes that solids levels are fixed throughout the storm, and no attempt is made to simulate sediment transport, erosion, or deposition. |
|
|
|
|
|
|
|
|
SPIDA calculates a solution to a set of one-dimensional partial differential equations relating discharge and water level in the links and nodes of the modelled network. The model thus computes flows and levels throughout the simulated network at each time step of the simulation. In any given time step, flooding occurs at a node when the water level is predicted to rise above the node's specified ground level. |
|
|
|
|
|
|
|
|
In reality, the relationship between water level, flooded surface |
|
|
|
|
|