Multiple domain evaluation of watershed hydrology models

Abstract

Abstract. Watershed scale models simulating hydrology and water quality have advanced rapidly in sophistication, process 5 representation, flexibility in model structure, and input data. Given the importance of these models to support decision-making for a wide range of environmental issues, the hydrology community is compelled to improve the metrics used to evaluate model performance. More targeted and comprehensive metrics will facilitate better and more efficient calibration and will help demonstrate that the model is useful for the intended purpose. Here we introduce a suite of new tools for model evaluation,packaged as an open-source Hydrologic Model Evaluation (HydroME) Toolbox. Specifically, we demonstrate the use of box 10 plots to illustrate the full distribution of common model performance metrics, such as R 2 , use of Euclidian distance, empirical Quantile-Quantile (Q-Q) plots and flow duration curves as simple metrics to identify and localize errors in model simulations.Further, we demonstrate the use of magnitude squared coherence to compare the frequency content between observed and modeled streamflow and wavelet coherence to localize frequency mismatches in time. We provide a rationale for a hierarchical selection of parameters to adjust during calibration and recommend that modelers progress from parameters with the most 15 uncertainty to the least uncertainty, namely starting with pure calibration parameters, followed by derived parameters, and finally measured parameters. We apply these techniques in the calibration and evaluation of models of two watersheds, the Le Sueur River Basin (2880 km 2 ) and Root River Basin (4300 km 2 ) in southern Minnesota, USA.

Development and Application of a Large‐Scale, Physically Based, Distributed Suspended Sediment Transport Model on the Fraser River Basin, British Columbia, Canada

“…However, computational and data limitations typically necessitate a coarse spatial resolution and a number of lumped or semidistributed parameters. Such limitations can restrict a model's ability to describe key processes and physical features such as river slope, which has been shown to be inaccurate when calculated directly from coarse resolution elevation data (Singh & Frevert, ; Wu et al, ), or lead to heavily parameterized models which are challenging to calibrate and suffer from equifinality (Beven, ; Kumarasamy & Belmont, ). To assess sediment load at a large scale, models should provide a means to resolve the influence of heterogeneity (variability) on sediment loads.…”

Solving water quality problems in agricultural landscapes: New approaches for these nonlinear, multiprocess, multiscale systems

“…These approaches begin to fill a critical gap in predictability. Prominent models used to predict the effects of agricultural practices at the watershed scale have advanced dramatically in sophistication, flexibility in model structure, and representation of management actions [ Santhi et al ., ; Chen and Wu , ; Gassman et al ., ; Kumarasamy and Belmont , ]. Yet they remain susceptible to deficiencies in methods to scale up from the field scale to large watersheds as well as representation of river channel dynamics, sediment storage, and transport processes, all critical components of the Minnesota River Basin system.…”

Review by A. Efstratiadis

“…In contrast, the recent (and not so recent) hydrological literature has widely discussed the shortcomings of the well-known NSE function, and the current approaches in hydrological calibration have been already incorporated most of the issues that the authors reveal (among many others, cf. Gupta et al, 2009;Ritter and Munoz-Carpena, 2013).…”