Hydrological Signature Representation of 7533 KGE Calibrated Conceptual Model Structures

Discussions calling for more rigorous evaluation practices for hydrologic models have recently increased. In addition to the widely used integral objective functions, hydrologic signatures are becoming common evaluation metrics for proving the suitability of hydrologic models for specific application purposes.

This work calibrates 7488 fixed conceptual model structures using KGE as an objective function. These structures range from a 1 to 3 storage model space previously used for an automatic model structure identification experiment. In this experiment we simultaneously calibrated the model structure (number of stores and flux equations) and its parameter values. Additionally, we calibrated 45 literature-based model structures (MARRMoT Toolbox) to extend the structural diversity in the analyzed models. We select well-performing models based on their KGE value (as is common practice) and analyze their performance using 12 selected hydrological runoff signatures. These signatures represent five aspects of the hydrological regime (magnitude, frequency, duration, rate of change, and timing). The large number of model structures, calibrated to the streamflow of 12 MOPEX catchments, allows general insight into how well common conceptual model structures can represent observed hydrological behavior evaluated by signatures.

Results show a general behavior of model structures calibrated to KGE to perform well in representing runoff ratio, mean discharge, the 95^th streamflow percentile, and the mean half-flow date. However, the analyzed conceptual model structures struggle to represent low flow and frequency signatures. When evaluated only for KGE, we can identify dominating model structures over all catchments. When evaluated for signatures, there are no model preferences over all analyzed catchments but different models seem to have their merits under specific conditions. These results support the need for ensuring model adequacy for a given task.