InfoWorks ICM 7.0 was recently released and amongst many new features, a new hydrological model is now available. Innovyze responded to requests from our user community and implemented the Clark Unit Hydrograph routing model.
Runoff routing models simulate stormwater flow along overland surfaces. ICM includes the most complete solutions in the market for runoff routing, as it can simulate runoff directly on the 2D overland mesh and on delineated catchments. The first approach applies the rainfall directly on the 2D mesh and runoff is simulated within the 2D model. The second approach based on catchments is the traditional way of modelling runoff routing and ICM includes the following models: Clark, Desbordes, Kadoya, Large Contributing Area, Non-Linear Routing, ReFH Unit Hydrograph, SCS Unit Hydrograph, Snyder Unit Hydrograph, SPRINT, SWMM, Unit Hydrograph, Wallingford Runoff Routing, Rational Routing. The Clark Unit Hydrograph model was the last one added to this list and was requested by several members of our modelling community, mainly from the USA.
The Clark model represents runoff routing based on two main processes: translation and attenuation. Translation is considered the movement of runoff within the catchment to its outlet and is represented with a time-area diagram, such as the one presented in the Figure 1. This diagram divides the fraction of the contributing area that discharge runoff from the catchment in a function of travel time since the beginning of the rainfall event.
Figure 1. Time-area diagram considered in Clark Unit Hydrograph model.
Attenuation is the reduction of runoff magnitude caused by storage and frictional forces in the catchment. It is derived with a linear reservoir representation based on the following equation:
where S is the reservoir storage, R is the constant linear reservoir parameter, and Q is outflow.
For more information about this model, please check the section “Clark Model” in the help file.
Probably you are now thinking “which model should I use?” There is not a unique answer and it will depend on each case. At first, the availability of data will dictate how complex the model can be. Data is required to feed model parameters and also to calibrate them. The model assumptions should also be considered to guarantee it is appropriate for the case. An example is the Wallingford runoff routing model that is applicable to small sub-catchments with areas up to 1ha. Another strong driver is the standard modelling approach of the region and specified guidance. The Clark Unit Hydrograph for example, has been applied in the USA so should be a great news for our modelling community in the USA. Of course in most of the situations, the user experience and preference dictates the model to use.