HydPy-ARMA

Many hydrologists use linear system approaches to solve flood routing problems approximately. Famous examples are the Unit Hydrograph approach and the Muskingum flood routing method. One can understand the Unit Hydrograph approach as a moving average process (MA), and the Muskingum method as a mixed autoregressive and moving average process (ARMA). Unit Hydrographs consist of an arbitrary number of MA coefficients. The Muskingum method defines three parameters, exactly two MA coefficients associated with the “new” and the “old” flow into the channel, and one AR coefficient associated with the “old” flow out of the channel.

The HydPy-ARMA base model supports implementing such methods in a generalised manner. Consider using one of its application models when you require a robust, computationally efficient, easily calibratable approach. However, taking nonlinear rating curves or backwater effects into account can be cumbersome or even impossible.

Available models:

• arma » HydPy-ARMA (base model)
• arma_rimorido » HydPy-ARMA-RIMO/RIDO (nonlinear routing by multiple ARMA processes)