The impact of a multi-criteria calibration on the performances of the DREAM model

Water resources observation and modelling are essential to better understand hydrological processes and improve water resource management. However, the reliability of hydrological simulation is strongly controlled by the quality and type of field observations used for the calibration and validation processes. Therefore, it is critical to develop proper strategies for model calibration and validation in order to reduce prediction uncertainties. Standard hydrological calibration relies mainly on the time series of total streamflow at the catchment outlet; nevertheless, this leads to a limited insight into the spatial behaviour of a river basin. In this work, we use simulations from the physically-based distributed DREAM model to discuss the importance of multi-criteria calibration to obtain consistent parameter sets. The calibration methodology exploits a physical based filter to decompose the streamflow times series in two time series referring to the surface component and the baseflow. Therefore, we adopted a multi-criteria calibration procedures which optimizes: (a) the total streamflow measured at the basin outlet (used as a reference study case); b) both the surface runoff and baseflow measured at the basin outlet; and (c) the combination the time series of the two components along with the annual water balance components. In addition, we also explored the use of a lumped parametrization against a spatial parametrization derived from the soil type characteristics of the river basin. In all cases, parameter optimization was carried out using an automatic calibration performed by a genetic algorithm (GA) tool. The study was carried out for two experimental catchments located in Basilicata and Campania regions (Southern Italy). The performed experiments showed that the inclusion of physical information during the calibration process results in a general improvement of model reliability.

This research is a part of iAqueduct project funded under the Water JPI 2018 Joint Call, Closing the Water Cycle Gap – on Sustainable Management of Water Resources - Water Works 2017.