Hierarchical mixture of experts and diagnostic modeling approach to reduce hydrologic model structural uncertainty
Abstract
In most water resources applications, a single model structure might be inadequate to capture the dynamic multi-scale interactions among different hydrological processes. Calibrating single models for dynamic catchments, where multiple dominant processes exist, can result in displacement of errors from structure to parameters, which in turn leads to over-correction and biased predictions. An alternative to a single model structure is to develop local expert structures that are effective in representing the dominant components of the hydrologic process and adaptively integrate them based on an indicator variable. In this study, the Hierarchical Mixture of Experts (HME) framework is applied to integrate expert model structures representing the different components of the hydrologic process. Various signature diagnostic analyses are used to assess the presence of multiple dominant processes and the adequacy of a single model, as well as to identify the structures of the expert models. The approaches are applied for two distinct catchments, the Guadalupe River (Texas) and the French Broad River (North Carolina) from the Model Parameter Estimation Experiment (MOPEX), using different structures of the HBV model. Furthermore, the results show that the HME approach has a better performance over the single model for the Guadalupe catchment, where multiple dominant processesmore »
- Authors:
-
- Washington State Univ., Richland, WA (United States)
- Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
- Publication Date:
- Research Org.:
- Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC), Biological and Environmental Research (BER); State of Washington Water Research Center
- OSTI Identifier:
- 1327118
- Alternate Identifier(s):
- OSTI ID: 1402274
- Report Number(s):
- PNNL-SA-113964
Journal ID: ISSN 0043-1397; KP1703030
- Grant/Contract Number:
- AC05-76RL01830; G11AP20113
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Water Resources Research
- Additional Journal Information:
- Journal Volume: 52; Journal Issue: 4; Journal ID: ISSN 0043-1397
- Publisher:
- American Geophysical Union (AGU)
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 54 ENVIRONMENTAL SCIENCES; diagnostic modeling; structural uncertainty; Hierarchical Mixture of Experts; model averaging; model adequacy
Citation Formats
Moges, Edom, Demissie, Yonas, and Li, Hong-Yi. Hierarchical mixture of experts and diagnostic modeling approach to reduce hydrologic model structural uncertainty. United States: N. p., 2016.
Web. doi:10.1002/2015WR018266.
Moges, Edom, Demissie, Yonas, & Li, Hong-Yi. Hierarchical mixture of experts and diagnostic modeling approach to reduce hydrologic model structural uncertainty. United States. https://doi.org/10.1002/2015WR018266
Moges, Edom, Demissie, Yonas, and Li, Hong-Yi. Fri .
"Hierarchical mixture of experts and diagnostic modeling approach to reduce hydrologic model structural uncertainty". United States. https://doi.org/10.1002/2015WR018266. https://www.osti.gov/servlets/purl/1327118.
@article{osti_1327118,
title = {Hierarchical mixture of experts and diagnostic modeling approach to reduce hydrologic model structural uncertainty},
author = {Moges, Edom and Demissie, Yonas and Li, Hong-Yi},
abstractNote = {In most water resources applications, a single model structure might be inadequate to capture the dynamic multi-scale interactions among different hydrological processes. Calibrating single models for dynamic catchments, where multiple dominant processes exist, can result in displacement of errors from structure to parameters, which in turn leads to over-correction and biased predictions. An alternative to a single model structure is to develop local expert structures that are effective in representing the dominant components of the hydrologic process and adaptively integrate them based on an indicator variable. In this study, the Hierarchical Mixture of Experts (HME) framework is applied to integrate expert model structures representing the different components of the hydrologic process. Various signature diagnostic analyses are used to assess the presence of multiple dominant processes and the adequacy of a single model, as well as to identify the structures of the expert models. The approaches are applied for two distinct catchments, the Guadalupe River (Texas) and the French Broad River (North Carolina) from the Model Parameter Estimation Experiment (MOPEX), using different structures of the HBV model. Furthermore, the results show that the HME approach has a better performance over the single model for the Guadalupe catchment, where multiple dominant processes are witnessed through diagnostic measures. Whereas, the diagnostics and aggregated performance measures prove that French Broad has a homogeneous catchment response, making the single model adequate to capture the response.},
doi = {10.1002/2015WR018266},
journal = {Water Resources Research},
number = 4,
volume = 52,
place = {United States},
year = {Fri Mar 04 00:00:00 EST 2016},
month = {Fri Mar 04 00:00:00 EST 2016}
}
Web of Science
Works referenced in this record:
Hierarchical Mixtures of Experts and the EM Algorithm
journal, March 1994
- Jordan, Michael I.; Jacobs, Robert A.
- Neural Computation, Vol. 6, Issue 2
Pitfalls and improvements in the joint inference of heteroscedasticity and autocorrelation in hydrological model calibration: Technical note
journal, July 2013
- Evin, Guillaume; Kavetski, Dmitri; Thyer, Mark
- Water Resources Research, Vol. 49, Issue 7
Maximum likelihood Bayesian averaging of spatial variability models in unsaturated fractured tuff: MAXIMUM LIKELIHOOD BAYESIAN MODEL AVERAGING
journal, May 2004
- Ye, Ming; Neuman, Shlomo P.; Meyer, Philip D.
- Water Resources Research, Vol. 40, Issue 5
Towards a Bayesian total error analysis of conceptual rainfall-runoff models: Characterising model error using storm-dependent parameters
journal, November 2006
- Kuczera, George; Kavetski, Dmitri; Franks, Stewart
- Journal of Hydrology, Vol. 331, Issue 1-2
Inference from Iterative Simulation Using Multiple Sequences
journal, November 1992
- Gelman, Andrew; Rubin, Donald B.
- Statistical Science, Vol. 7, Issue 4
Comparative diagnostic analysis of runoff generation processes in Oklahoma DMIP2 basins: The Blue River and the Illinois River
journal, February 2012
- Li, Hongyi; Sivapalan, Murugesu; Tian, Fuqiang
- Journal of Hydrology, Vol. 418-419
Teaching hydrological modeling with a user-friendly catchment-runoff-model software package
journal, January 2012
- Seibert, J.; Vis, M. J. P.
- Hydrology and Earth System Sciences, Vol. 16, Issue 9
Treatment of input uncertainty in hydrologic modeling: Doing hydrology backward with Markov chain Monte Carlo simulation: FORCING DATA ERROR USING MCMC SAMPLING
journal, December 2008
- Vrugt, Jasper A.; ter Braak, Cajo J. F.; Clark, Martyn P.
- Water Resources Research, Vol. 44, Issue 12
Detecting human interferences to low flows through base flow recession analysis: DETECTING HUMAN INTERFERENCES TO LOW FLOWS
journal, July 2009
- Wang, Dingbao; Cai, Ximing
- Water Resources Research, Vol. 45, Issue 7
Accelerating Markov Chain Monte Carlo Simulation by Differential Evolution with Self-Adaptive Randomized Subspace Sampling
journal, January 2009
- Vrugt, J. A.; ter Braak, C. J. F.; Diks, C. G. H.
- International Journal of Nonlinear Sciences and Numerical Simulation, Vol. 10, Issue 3
Modeling the catchment via mixtures: Issues of model specification and validation: MODELING THE CATCHMENT VIA MIXTURES
journal, November 2006
- Marshall, Lucy; Sharma, Ashish; Nott, David
- Water Resources Research, Vol. 42, Issue 11
Towards dynamic catchment modelling: a Bayesian hierarchical mixtures of experts framework
journal, January 2007
- Marshall, Lucy; Nott, David; Sharma, Ashish
- Hydrological Processes, Vol. 21, Issue 7
Multi-model ensemble hydrologic prediction using Bayesian model averaging
journal, May 2007
- Duan, Qingyun; Ajami, Newsha K.; Gao, Xiaogang
- Advances in Water Resources, Vol. 30, Issue 5
Bayesian recursive parameter estimation for hydrologic models
journal, October 2001
- Thiemann, M.; Trosset, M.; Gupta, H.
- Water Resources Research, Vol. 37, Issue 10
Bayesian analysis of input uncertainty in hydrological modeling: 2. Application: INPUT UNCERTAINTY IN HYDROLOGY, 2
journal, March 2006
- Kavetski, Dmitri; Kuczera, George; Franks, Stewart W.
- Water Resources Research, Vol. 42, Issue 3
Specifying a hierarchical mixture of experts for hydrologic modeling: Gating function variable selection: Specifying A Hierarchical Mixture of Experts
journal, May 2013
- Jeremiah, Erwin; Marshall, Lucy; Sisson, Scott A.
- Water Resources Research, Vol. 49, Issue 5
Towards the characterization of streamflow simulation uncertainty through multimodel ensembles
journal, October 2004
- Georgakakos, Konstantine P.; Seo, Dong-Jun; Gupta, Hoshin
- Journal of Hydrology, Vol. 298, Issue 1-4
Pursuing the method of multiple working hypotheses for hydrological modeling: HYPOTHESIS TESTING IN HYDROLOGY
journal, September 2011
- Clark, Martyn P.; Kavetski, Dmitri; Fenicia, Fabrizio
- Water Resources Research, Vol. 47, Issue 9
Toward improved calibration of hydrologic models: Multiple and noncommensurable measures of information
journal, April 1998
- Gupta, Hoshin Vijai; Sorooshian, Soroosh; Yapo, Patrice Ogou
- Water Resources Research, Vol. 34, Issue 4
The future of distributed models: Model calibration and uncertainty prediction
journal, July 1992
- Beven, Keith; Binley, Andrew
- Hydrological Processes, Vol. 6, Issue 3
Decomposition of the mean squared error and NSE performance criteria: Implications for improving hydrological modelling
journal, October 2009
- Gupta, Hoshin V.; Kling, Harald; Yilmaz, Koray K.
- Journal of Hydrology, Vol. 377, Issue 1-2
A formal likelihood function for parameter and predictive inference of hydrologic models with correlated, heteroscedastic, and non‐Gaussian errors
journal, October 2010
- Schoups, Gerrit; Vrugt, Jasper A.
- Water Resources Research, Vol. 46, Issue 10
An evaluation of the impact of model structure on hydrological modelling uncertainty for streamflow simulation
journal, October 2004
- Butts, Michael B.; Payne, Jeffrey T.; Kristensen, Michael
- Journal of Hydrology, Vol. 298, Issue 1-4
Consistency between hydrological models and field observations: linking processes at the hillslope scale to hydrological responses at the watershed scale
journal, January 2009
- Clark, M. P.; Rupp, D. E.; Woods, R. A.
- Hydrological Processes, Vol. 23, Issue 2
Large-sample hydrology: a need to balance depth with breadth
journal, January 2014
- Gupta, H. V.; Perrin, C.; Blöschl, G.
- Hydrology and Earth System Sciences, Vol. 18, Issue 2
Using Bayesian Model Averaging to Calibrate Forecast Ensembles
journal, May 2005
- Raftery, Adrian E.; Gneiting, Tilmann; Balabdaoui, Fadoua
- Monthly Weather Review, Vol. 133, Issue 5, p. 1155-1174
Towards reduced uncertainty in conceptual rainfall-runoff modelling: dynamic identifiability analysis
journal, January 2003
- Wagener, T.; McIntyre, N.; Lees, M. J.
- Hydrological Processes, Vol. 17, Issue 2
Spillway design floods in Sweden: I. New guidelines
journal, October 1992
- BergstrÖM, Sten; Harlin, Joakim; LindstrÖM, GÖRan
- Hydrological Sciences Journal, Vol. 37, Issue 5
A Bayesian perspective on input uncertainty in model calibration: Application to hydrological model “abc”: INPUT UNCERTAINTY IN MODEL CALIBRATION
journal, July 2006
- Huard, David; Mailhot, Alain
- Water Resources Research, Vol. 42, Issue 7
Prediction uncertainty of conceptual rainfall-runoff models caused by problems in identifying model parameters and structure
journal, October 1999
- Uhlenbrook, Stefan; Seibert, Jan; Leibundgut, Christian
- Hydrological Sciences Journal, Vol. 44, Issue 5
Treatment of uncertainty using ensemble methods: Comparison of sequential data assimilation and Bayesian model averaging: SEQUENTIAL DATA ASSIMILATION
journal, January 2007
- Vrugt, Jasper A.; Robinson, Bruce A.
- Water Resources Research, Vol. 43, Issue 1
Reconciling theory with observations: elements of a diagnostic approach to model evaluation
journal, August 2008
- Gupta, Hoshin V.; Wagener, Thorsten; Liu, Yuqiong
- Hydrological Processes, Vol. 22, Issue 18
Development of a formal likelihood function for improved Bayesian inference of ephemeral catchments: DEVELOPMENT OF A FORMAL LIKELIHOOD FUNCTION
journal, December 2010
- Smith, Tyler; Sharma, Ashish; Marshall, Lucy
- Water Resources Research, Vol. 46, Issue 12
Functional approach to exploring climatic and landscape controls of runoff generation: 1. Behavioral constraints on runoff volume
journal, December 2014
- Li, Hong-Yi; Sivapalan, Murugesu; Tian, Fuqiang
- Water Resources Research, Vol. 50, Issue 12
Joint application of event-based calibration and dynamic identifiability analysis in rainfall–runoff modelling: implications for model parametrisation
journal, October 2008
- Cullmann, Johannes; Wriedt, Gunter
- Journal of Hydroinformatics, Vol. 10, Issue 4
Benchmarking observational uncertainties for hydrology: rainfall, river discharge and water quality: BENCHMARKING OBSERVATIONAL UNCERTAINTIES FOR HYDROLOGY
journal, June 2012
- McMillan, Hilary; Krueger, Tobias; Freer, Jim
- Hydrological Processes, Vol. 26, Issue 26
Predicting space-time variability of hourly streamflow and the role of climate seasonality: Mahurangi Catchment, New Zealand
journal, January 2003
- Atkinson, S. E.; Sivapalan, M.; Viney, N. R.
- Hydrological Processes, Vol. 17, Issue 11
When are multiobjective calibration trade-offs in hydrologic models meaningful?: MEANINGFUL MULTIOBJECTIVE TRADE-OFFS
journal, March 2012
- Kollat, J. B.; Reed, P. M.; Wagener, T.
- Water Resources Research, Vol. 48, Issue 3
The influence of lateral snow redistribution processes on snow melt and sublimation in alpine regions
journal, March 2012
- Bernhardt, M.; Schulz, K.; Liston, G. E.
- Journal of Hydrology, Vol. 424-425
Understanding predictive uncertainty in hydrologic modeling: The challenge of identifying input and structural errors: IDENTIFIABILITY OF INPUT AND STRUCTURAL ERRORS
journal, May 2010
- Renard, Benjamin; Kavetski, Dmitri; Kuczera, George
- Water Resources Research, Vol. 46, Issue 5
Uncertainty in river discharge observations: a quantitative analysis
journal, January 2009
- Di Baldassarre, G.; Montanari, A.
- Hydrology and Earth System Sciences, Vol. 13, Issue 6
Hydrological field data from a modeller's perspective: Part 1. Diagnostic tests for model structure
journal, November 2010
- McMillan, Hilary K.; Clark, Martyn P.; Bowden, William B.
- Hydrological Processes, Vol. 25, Issue 4
An introduction to the hydrological modelling system PREVAH and its pre- and post-processing-tools
journal, October 2009
- Viviroli, D.; Zappa, M.; Gurtz, J.
- Environmental Modelling & Software, Vol. 24, Issue 10
A framework to assess the realism of model structures using hydrological signatures
journal, January 2013
- Euser, T.; Winsemius, H. C.; Hrachowitz, M.
- Hydrology and Earth System Sciences, Vol. 17, Issue 5
Spatial variability of hydrological processes and model structure diagnostics in a 50 km 2 catchment : SPATIAL VARIABILITY OF HYDROLOGICAL PROCESSES AND MODEL DIAGNOSTICS
journal, August 2013
- McMillan, Hilary; Gueguen, Myriam; Grimon, Elisabeth
- Hydrological Processes, Vol. 28, Issue 18
Interpretation of runoff processes in hydrological modelling-experience from the HBV approach: Interpretation of Runoff Processes in Hydrological Modelling
journal, May 2015
- Bergström, Sten; Lindström, Göran
- Hydrological Processes, Vol. 29, Issue 16
Unraveling uncertainties in hydrologic model calibration: Addressing the problem of compensatory parameters
journal, January 2006
- Clark, Martyn P.; Vrugt, Jasper A.
- Geophysical Research Letters, Vol. 33, Issue 6
Process consistency in models: The importance of system signatures, expert knowledge, and process complexity
journal, September 2014
- Hrachowitz, M.; Fovet, O.; Ruiz, L.
- Water Resources Research, Vol. 50, Issue 9
Methods for combining the outputs of different rainfall–runoff models
journal, October 1997
- Shamseldin, Asaad Y.; O'Connor, Kieran M.; Liang, G. C.
- Journal of Hydrology, Vol. 197, Issue 1-4
Framework for Understanding Structural Errors (FUSE): A modular framework to diagnose differences between hydrological models: DIFFERENCES BETWEEN HYDROLOGICAL MODELS
journal, August 2008
- Clark, Martyn P.; Slater, Andrew G.; Rupp, David E.
- Water Resources Research, Vol. 44, Issue 12
Comparison of joint versus postprocessor approaches for hydrological uncertainty estimation accounting for error autocorrelation and heteroscedasticity
journal, March 2014
- Evin, Guillaume; Thyer, Mark; Kavetski, Dmitri
- Water Resources Research, Vol. 50, Issue 3
Modeling residual hydrologic errors with Bayesian inference
journal, September 2015
- Smith, Tyler; Marshall, Lucy; Sharma, Ashish
- Journal of Hydrology, Vol. 528
Improving model structure and reducing parameter uncertainty in conceptual water balance models through the use of auxiliary data: IMPROVING MODEL STRUCTURE THROUGH AUXILIARY DATA
journal, January 2007
- Son, Kyongho; Sivapalan, Murugesu
- Water Resources Research, Vol. 43, Issue 1
On the scale problem in hydrological modelling
journal, November 1998
- Bergström, Sten; Graham, L. Phil
- Journal of Hydrology, Vol. 211, Issue 1-4
Works referencing / citing this record:
A Prior Estimation of the Spatial Distribution Parameter of Soil Moisture Storage Capacity Using Satellite-Based Root-Zone Soil Moisture Data
journal, November 2019
- Tian, Yifei; Xiong, Lihua; Xiong, Bin
- Remote Sensing, Vol. 11, Issue 21