Estimation of Models for the Upper Part of Murray River with Flow Dependent Parameters

Abstract

Data-based modelling is an important tool for efficient planning and management of rivers. However, linear data-based models often only perform well in a specific operating range. The larger the operating range is, the better the model is considered for prediction and control. An increase in the operating range can be achieved by incorporating non-linearities in the model. In this paper we used Data-Based Mechanistic (DBM) approach (Young (2011)) to derive models of the upper part of Murray River in Australia. The approach can identify static non-linearities in the system, and we used it to see how significant the non-linearities in the river model are. We analysed models obtained from historical data spanning a decade. In the data considered we could not find any strong indication that significant non-linearities were present.