Simultaneously determining global sensitivities of model parameters and model structure

Abstract. Model structure uncertainty is known to be one of the three main sources of hydrologic model uncertainty along with input and parameter uncertainty. Some recent hydrological modeling frameworks address model structure uncertainty by supporting multiple options for representing hydrological processes. It is, however, still unclear how best to analyze structural sensitivity using these frameworks. In this work, we apply the extended Sobol' sensitivity analysis (xSSA) method that operates on grouped parameters rather than individual parameters. The method can estimate not only traditional model parameter sensitivities but is also able to provide measures of the sensitivities of process options (e.g., linear vs. non-linear storage) and sensitivities of model processes (e.g., infiltration vs. baseflow) with respect to a model output. Key to the xSSA method's applicability to process option and process sensitivity is the novel introduction of process option weights in the Raven hydrological modeling framework. The method is applied to both artificial benchmark models and a watershed model built with the Raven framework. The results show that (1) the xSSA method provides sensitivity estimates consistent with those derived analytically for individual as well as grouped parameters linked to model structure. (2) The xSSA method with process weighting is computationally less expensive than the alternative aggregate sensitivity analysis approach performed for the exhaustive set of structural model configurations, with savings of 81.9 % for the benchmark model and 98.6 % for the watershed case study. (3) The xSSA method applied to the hydrologic case study analyzing simulated streamflow showed that model parameters adjusting forcing functions were responsible for 42.1 % of the overall model variability, while surface processes cause 38.5 % of the overall model variability in a mountainous catchment; such information may readily inform model calibration and uncertainty analysis. (4) The analysis of time-dependent process sensitivities regarding simulated streamflow is a helpful tool for understanding model internal dynamics over the course of the year.

Download Full-text

Simultaneously Determining Global Sensitivities of Model Parameters and Model Structure

10.5194/hess-2020-215 ◽

2020 ◽

Author(s):

Juliane Mai ◽

James R. Craig ◽

Bryan A. Tolson

Keyword(s):

Sensitivity Analysis ◽

Hydrological Modeling ◽

Structural Model ◽

Hydrologic Model ◽

Model Parameters ◽

Model Structure ◽

Modeling Framework ◽

Model Structure Uncertainty ◽

Model Variability

Abstract. Model structure uncertainty is known to be one of the three main sources of hydrologic model uncertainty along with input and parameter uncertainty. Some recent hydrological modeling frameworks address model structure uncertainty by supporting multiple options for representing hydrological processes. It is, however, still unclear how best to analyze structural sensitivity using these frameworks. In this work, we apply an Extended Sobol' Sensitivity Analysis (xSSA) method that operates on grouped parameters rather than individual parameters. The method can estimate not only traditional model parameter sensitivities but is also able to provide measures of the sensitivities of process options (e.g., linear vs. non-linear storage) and sensitivities of model processes (e.g., infiltration vs. baseflow) with respect to a model output. Key to the xSSA method's applicability to process option and process sensitivity is the novel introduction of process option weights in the Raven hydrological modeling framework. The method is applied to both artificial benchmark models and a watershed model built with the Raven framework. The results show that: (1) The xSSA method provides sensitivity estimates consistent with those derived analytically for individual as well as grouped parameters linked to model structure. (2) The xSSA method with process weighting is computationally less expensive than the alternative aggregate sensitivity analysis approach performed for the exhaustive set of structural model configurations, with savings of 81.9 % for the benchmark model and 98.6 % for the watershed case study. (3) The xSSA method applied to the hydrologic case study analyzing simulated streamflow showed that model parameters adjusting forcing functions were responsible for 42.1 % of the overall model variability while surface processes cause 38.5 % of the overall model variability in a mountainous catchment; such information may readily inform model calibration. (4) The analysis of time dependent process sensitivities regarding simulated streamflow is a helpful tool to understand model internal dynamics over the course of the year.

Download Full-text

Identifying sensitivities in flood frequency analyses using a stochastic hydrologic modeling system

Hydrology and Earth System Sciences ◽

10.5194/hess-25-5603-2021 ◽

2021 ◽

Vol 25 (10) ◽

pp. 5603-5621

Author(s):

Andrew J. Newman ◽

Amanda G. Stone ◽

Manabendra Saharia ◽

Kathleen D. Holman ◽

Nans Addor ◽

...

Keyword(s):

Hydrologic Modeling ◽

Initial Conditions ◽

Hydrologic Model ◽

Flood Frequency ◽

Model Parameters ◽

Model Structure ◽

Structure Model ◽

Return Periods ◽

Modeling Framework ◽

Flood Generation

Abstract. This study employs a stochastic hydrologic modeling framework to evaluate the sensitivity of flood frequency analyses to different components of the hydrologic modeling chain. The major components of the stochastic hydrologic modeling chain, including model structure, model parameter estimation, initial conditions, and precipitation inputs were examined across return periods from 2 to 100 000 years at two watersheds representing different hydroclimates across the western USA. A total of 10 hydrologic model structures were configured, calibrated, and run within the Framework for Understanding Structural Errors (FUSE) modular modeling framework for each of the two watersheds. Model parameters and initial conditions were derived from long-term calibrated simulations using a 100 member historical meteorology ensemble. A stochastic event-based hydrologic modeling workflow was developed using the calibrated models in which millions of flood event simulations were performed for each basin. The analysis of variance method was then used to quantify the relative contributions of model structure, model parameters, initial conditions, and precipitation inputs to flood magnitudes for different return periods. Results demonstrate that different components of the modeling chain have different sensitivities for different return periods. Precipitation inputs contribute most to the variance of rare floods, while initial conditions are most influential for more frequent events. However, the hydrological model structure and structure–parameter interactions together play an equally important role in specific cases, depending on the basin characteristics and type of flood metric of interest. This study highlights the importance of critically assessing model underpinnings, understanding flood generation processes, and selecting appropriate hydrological models that are consistent with our understanding of flood generation processes.

Download Full-text

Spatial Pattern Oriented Multi-Criteria Sensitivity Analysis of a Distributed Hydrologic Model

10.20944/preprints201808.0209.v1 ◽

2018 ◽

Author(s):

Mehmet Cüneyd Demirel ◽

Julian Koch ◽

Gorka Mendiguren ◽

Simon Stisen

Keyword(s):

Sensitivity Analysis ◽

Spatial Variability ◽

Spatial Pattern ◽

Spatial Patterns ◽

Performance Measures ◽

Sampling Strategy ◽

Behavioral Model ◽

Hydrologic Model ◽

Actual Evapotranspiration ◽

Model Parameters

Hydrologic models are conventionally constrained and evaluated using point measurements of streamflow, which represents an aggregated catchment measure. As a consequence of this single objective focus, model parametrization and model parameter sensitivity are typically not reflecting other aspects of catchment behavior. Specifically for distributed models, the spatial pattern aspect is often overlooked. Our paper examines the utility of multiple performance measures in a spatial sensitivity analysis framework to determine the key parameters governing the spatial variability of predicted actual evapotranspiration (AET). Latin hypercube one-at-a-time (LHS-OAT) sampling strategy with multiple initial parameter sets was applied using the mesoscale hydrologic model (mHM) and a total of 17 model parameters were identified as sensitive. The results indicate different parameter sensitivities for different performance measures focusing on temporal hydrograph dynamics and spatial variability of actual evapotranspiration. While spatial patterns were found to be sensitive to vegetation parameters, streamflow dynamics were sensitive to pedo-transfer function (PTF) parameters. Above all, our results show that behavioral model definition based only on streamflow metrics in the generalized likelihood uncertainty estimation (GLUE) type methods require reformulation by incorporating spatial patterns into the definition of threshold values to reveal robust hydrologic behavior in the analysis.

Download Full-text

Comparison between dynamic and static sensitivity analysis approaches for impact assessment of different potential evapotranspiration methods on hydrological models performance

Journal of Hydrometeorology ◽

10.1175/jhm-d-20-0192.1 ◽

2021 ◽

Author(s):

Rodric Mérimé Nonki ◽

André Lenouo ◽

Christopher J. Lennard ◽

Raphael M. Tshimanga ◽

Clément Tchawoua

Keyword(s):

Sensitivity Analysis ◽

Impact Assessment ◽

Hydrological Modeling ◽

Potential Evapotranspiration ◽

Model Performance ◽

Estimation Methods ◽

Model Parameters ◽

Hydrological Models ◽

Sensitivity Approach ◽

Static Sensitivity

AbstractPotential Evapotranspiration (PET) plays a crucial role in water management, including irrigation systems design and management. It is an essential input to hydrological models. Direct measurement of PET is difficult, time-consuming and costly, therefore a number of different methods are used to compute this variable. This study compares the two sensitivity analysis approaches generally used for PET impact assessment on hydrological model performance. We conducted the study in the Upper Benue River Basin (UBRB) located in northern Cameroon using two lumped-conceptual rainfall-runoff models and nineteen PET estimation methods. A Monte-Carlo procedure was implemented to calibrate the hydrological models for each PET input while considering similar objective functions. Although there were notable differences between PET estimation methods, the hydrological models performance was satisfactory for each PET input in the calibration and validation periods. The optimized model parameters were significantly affected by the PET-inputs, especially the parameter responsible to transform PET into actual ET. The hydrological models performance was insensitive to the PET input using a dynamic sensitivity approach, while he was significantly affected using a static sensitivity approach. This means that the over-or under-estimation of PET is compensated by the model parameters during the model recalibration. The model performance was insensitive to the rescaling PET input for both dynamic and static sensitivities approaches. These results demonstrate that the effect of PET input to model performance is necessarily dependent on the sensitivity analysis approach used and suggest that the dynamic approach is more effective for hydrological modeling perspectives.

Download Full-text

Spatial disaggregation of bias-corrected GCM precipitation for improved hydrologic simulation: Ping River Basin, Thailand

Hydrology and Earth System Sciences ◽

10.5194/hess-11-1373-2007 ◽

2007 ◽

Vol 11 (4) ◽

pp. 1373-1390 ◽

Cited By ~ 131

Author(s):

D. Sharma ◽

A. Das Gupta ◽

M. S. Babel

Keyword(s):

River Basin ◽

Bias Correction ◽

Hydrological Modeling ◽

Correction Method ◽

Hydrologic Model ◽

Global Climate Models ◽

Model Parameters ◽

Bias Correction Method ◽

Spatial Disaggregation ◽

Grid Nodes

Abstract. Global Climate Models (GCMs) precipitation scenarios are often characterized by biases and coarse resolution that limit their direct application for basin level hydrological modeling. Bias-correction and spatial disaggregation methods are employed to improve the quality of ECHAM4/OPYC SRES A2 and B2 precipitation for the Ping River Basin in Thailand. Bias-correction method, based on gamma-gamma transformation, is applied to improve the frequency and amount of raw GCM precipitation at the grid nodes. Spatial disaggregation model parameters (β,σ2), based on multiplicative random cascade theory, are estimated using Mandelbrot-Kahane-Peyriere (MKP) function at q=1 for each month. Bias-correction method exhibits ability of reducing biases from the frequency and amount when compared with the computed frequency and amount at grid nodes based on spatially interpolated observed rainfall data. Spatial disaggregation model satisfactorily reproduces the observed trend and variation of average rainfall amount except during heavy rainfall events with certain degree of spatial and temporal variations. Finally, the hydrologic model, HEC-HMS, is applied to simulate the observed runoff for upper Ping River Basin based on the modified GCM precipitation scenarios and the raw GCM precipitation. Precipitation scenario developed with bias-correction and disaggregation provides an improved reproduction of basin level runoff observations.

Download Full-text

Interpreting the Estimates from the Full VECH Model with Asymmetry: The Case of US and Canadian Equity Prices

Accounting and Finance Research ◽

10.5430/afr.v6n4p236 ◽

2017 ◽

Vol 6 (4) ◽

pp. 236

Author(s):

Chikashi Tsuji

Keyword(s):

Parameter Estimates ◽

Model Parameters ◽

Model Structure ◽

Equity Prices ◽

Related Research ◽

Asymmetric Effects ◽

Equity Index ◽

Autoregressive Conditional Heteroscedasticity ◽

Full Vector

This paper attempts to derive careful interpretation of the parameter estimates from one of the multivariate generalized autoregressive conditional heteroscedasticity (MGARCH) models, the full vector-half (VECH) model with asymmetric effects. We also consider and interpret the parameter estimates from a case study of US and Canadian equity index returns by applying this model. More specifically, we firstly inspect the model formula and derive general interpretation of the model parameters. We consider this is particularly useful for understanding not only the full VECH model structure but also similar MGARCH models. After the general considerations, we also interpret the case results that are derived from our application of the full VECH model to US and Canadian equity index returns. We consider that these concrete illustrations are also very helpful for future related research.

Download Full-text

Runoff Predicting and Variation Analysis in Upper Ganjiang Basin under Projected Climate Changes

Sustainability ◽

10.3390/su11215885 ◽

2019 ◽

Vol 11 (21) ◽

pp. 5885 ◽

Cited By ~ 3

Author(s):

Chao Deng ◽

Weiguang Wang

Keyword(s):

Resource Management ◽

Water Resource ◽

Water Resource Management ◽

Hydrological Modeling ◽

Hydrologic Model ◽

Global Climate Models ◽

Balance Model ◽

Model Parameters ◽

The Mean ◽

Parameter Approach

Catchment runoff is significantly affected by climate condition changes. Predicting the runoff and analyzing its variations under future climates play a vital role in water security, water resource management, and the sustainable development of the catchment. In traditional hydrological modeling, fixed model parameters are usually used to transfer the global climate models (GCMs) to runoff, while the hydrologic model parameters may be time-varying. It is more appropriate to use the time-variant parameter for runoff modeling. This is achieved by incorporating the time-variant parameter approach into a two-parameter water balance model (TWBM) through the construction of time-variant parameter functions based on the identified catchment climate indicators. Using the Ganjiang Basin with an outlet of the Dongbei Hydrological Station as the study area, we developed time-variant parameter scenarios of the TWBM model and selected the best-performed parameter functions to predict future runoff and analyze its variations under the climate model projection of the BCC-CSM1.1(m). To synthetically assess the model performance improvements using the time-variant parameter approach, an index Δ was developed by combining the Nash–Sutcliffe efficiency, the volume error, the Box–Cox transformed root-mean-square error, and the Kling–Gupta efficiency with equivalent weight. The results show that the TWBM model with time-variant C (evapotranspiration parameter) and SC (water storage capacity of catchment), where growing and non-growing seasons are considered for C, outperformed the model with constant parameters with a Δ value of approximately 5% and 10% for the calibration and validation periods, respectively. The mean annual values of runoff predictions under the four representative concentration pathways (RCPs) exhibited a decreasing trend over the future three decades (2021–2050) when compared to the runoff simulations in the baseline period (1982–2011), where the values were about −9.9%, −19.5%, −16.6%, and −11.4% for the RCP2.6, RCP4.5, RCP6.0, and RCP8.5, respectively. The decreasing trend of future precipitation exerts impacts on runoff decline. Generally, the mean monthly changes of runoff predictions showed a decreasing trend from January to August for almost all of the RCPs, while an increasing trend existed from September to November, along with fluctuations among different RCPs. This study can provide beneficial references to comprehensively understand the impacts of climate change on runoff prediction and thus improve the regional strategy for future water resource management.

Download Full-text