Multicriteria sensitivity analysis as a diagnostic tool for understanding model behaviour and characterizing model uncertainty

Abstract

Complex hydrological models are being increasingly used nowadays for many purposes such as studying the impact of climate and land‐use change on water resources. However, building a high‐fidelity model, particularly at large scales, remains a challenging task, due to complexities in model functioning and behaviour and uncertainties in model structure, parameterization, and data. Global sensitivity analysis (GSA), which characterizes how the variation in the model response is attributed to variations in its input factors (e.g., parameters and forcing data), provides an opportunity to enhance the development and application of these complex models. In this paper, we advocate using GSA as an integral part of the modelling process by discussing its capabilities as a tool for diagnosing model structure and detecting potential defects, identifying influential factors, characterizing uncertainty, and selecting calibration parameters. Accordingly, we conduct a comprehensive GSA of a complex land surface–hydrology model, Modélisation Environmentale–Surface et Hydrologie (MESH), which combines the Canadian land surface scheme with a hydrological routing component, WATROUTE. Various GSA experiments are carried out using a new technique, called Variogram Analysis of Response Surfaces, for alternative hydroclimatic conditions in Canada using multiple criteria, various model configurations, and a full set of model parameters. Results from this study reveal that, in addition to different hydroclimatic conditions and SA criteria, model configurations can also have a major impact on the assessment of sensitivity. GSA can identify aspects of the model internal functioning that are counter‐intuitive and thus help the modeller to diagnose possible model deficiencies and make recommendations for improving development and application of the model. As a specific outcome of this work, a list of the most influential parameters for the MESH model is developed. This list, along with some specific recommendations, is expected to assist the wide community of MESH and Canadian land surface scheme users, to enhance their modelling applications.

Revisiting the Basis of Sensitivity Analysis for Dynamical Earth System Models

“…If the fidelity of the model is poor (see Clark et al, ), then the GSA will likely produce results that are artifacts and cannot be trusted. Of course, if the GSA provides results that run clearly counter to our intuitive understanding of the system, we may be able to use this information as a diagnostic tool to help us figure out what model improvements are necessary (Haghnegahdar et al, ). In our case, we did a crude preliminary analysis to characterize streamflow performance of the HBV‐SASK model for the Oldman and Banff basins, by randomly sampling several thousand parameter locations across the feasible space and checking the frequency distributions of normalized MSEs so obtained ( NMSE = MSE / Var ( Q Obs )).…”

Impacts of future climate on the hydrology of a northern headwaters basin and its implications for a downstream deltaic ecosystem

“…There is an ongoing effort in Canada, led by a team of researchers from ECCC and the University of Saskatchewan, to further develop, improve, and apply the MESH modelling system for most of Canada. These improvements are aimed to strengthen MESH's ability to simulate hydrological processes adequately under current and changing environmental conditions in Canada and across the globe (Haghnegahdar, Razavi, Yassin, & Wheater, ). MESH is currently being used within many large research projects, such as the Drought Research Initiative (http://www.meteo.mcgill.ca/dri/), Changing Cold Regions Network (http://www.ccrnetwork.ca/), and Global Water Futures (https://gwf.usask.ca/).…”

“…There is an ongoing effort in Canada, led by a team of researchers from ECCC and the University of Saskatchewan, to further develop, improve, and apply the MESH modelling system for most of Canada. These improvements are aimed to strengthen MESH's ability to simulate hydrological processes adequately under current and changing environmental conditions in Canada and across the globe (Haghnegahdar, Razavi, Yassin, & Wheater, 2017 Verseghy, 1991;Verseghy, McFarlane, & Lazare, 1993) or the soil, vegetation, and snow scheme (SVS; Husain et al, 2016). The lateral soil (subsurface) and surface water movement are simulated by either WATROF (Soulis, Snelgrove, Kouwen, Seglenieks, & Verseghy, 2000) or PDMROF (Mekonnen et al, 2014).…”

“…Three families of parameters, that is, vegetation parameters, lateral flow parameters, and routing parameters, were calibrated based on previous multicriteria sensitivity analyses of MESH parameters (e.g., Haghnegahdar et al, 2017;Yassin et al, 2017); see Table 2 for the list of parameters and their calibration ranges. The initial parameter values were adapted from a previous study of the Peace River by Rokaya et al (2019b initial conditions on simulation outputs.…”

A comprehensive sensitivity and uncertainty analysis for discharge and nitrate-nitrogen loads involving multiple discrete model inputs under future changing conditions

“…Environmental model applications are usually less computationally expensive and allow a more extensive GSA, illustrated in many environmental modeling studies (e.g. Guse et al, 2016b;Haghnegahdar et al, 2017;Massmann and Holzmann, 2015;Razavi and Gupta, 2016b;Sarrazin et al, 2016).…”