Explicitly integrating parameter, input, and structure uncertainties into Bayesian Neural Networks for probabilistic hydrologic forecasting

Summary Estimating uncertainty of hydrologic forecasting is valuable to water resources and other relevant decision making processes. Recently, Bayesian Neural Networks (BNNs) have been proved powerful tools for quantifying uncertainty of streamflow forecasting. In this study, we propose a Markov Chain Monte Carlo (MCMC) framework (BNN-PIS) to incorporate the uncertainties associated with parameters, inputs, and structures into BNNs. This framework allows the structure of the neural networks to change by removing or adding connections between neurons and enables scaling of input data by using rainfall multipliers. The results show that the new BNNs outperform BNNs that only consider uncertainties associated with parameters and model structures. Critical evaluation of posterior distribution of neural network weights, number of effective connections, rainfall multipliers, and hyper-parameters shows that the assumptions held in our BNNs are not well supported. Further understanding of characteristics of and interactions among different uncertainty sources is expected to enhance the application of neural networks for uncertainty analysis of hydrologic forecasting.

[1]  Holger R. Maier,et al.  Bayesian model selection applied to artificial neural networks used for water resources modeling , 2008 .

[2]  Jouko Lampinen,et al.  Bayesian approach for neural networks--review and case studies , 2001, Neural Networks.

[3]  A. Montanari,et al.  Uncertainty in river discharge observations: a quantitative analysis , 2009 .

[4]  George Kuczera,et al.  Understanding predictive uncertainty in hydrologic modeling: The challenge of identifying input and structural errors , 2010 .

[5]  George Kuczera,et al.  Bayesian analysis of input uncertainty in hydrological modeling: 1. Theory , 2006 .

[6]  Shouhong Wang.  The unpredictability of standard back propagation neural networks in classification applications , 1995 .

[7]  N. Null.  Artificial Neural Networks in Hydrology. I: Preliminary Concepts , 2000 .

[8]  Keith Beven,et al.  A manifesto for the equifinality thesis , 2006 .

[9]  Yen-Ming Chiang,et al.  Dynamic ANN for precipitation estimation and forecasting from radar observations , 2007 .

[10]  K. P. Sudheer,et al.  Methods used for the development of neural networks for the prediction of water resource variables in river systems: Current status and future directions , 2010, Environ. Model. Softw..

[11]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[12]  Christian W. Dawson,et al.  Hydrological modelling using artificial neural networks , 2001 .

[13]  Carmen Cirincione,et al.  Municipal Government Revenue Forecasting: Issues of Method and Data , 1999 .

[14]  R. K. Hubbard,et al.  Little River Experimental Watershed database , 2007 .

[15]  H. Hersbach.  Decomposition of the Continuous Ranked Probability Score for Ensemble Prediction Systems , 2000 .

[16]  D. Kavetski,et al.  Towards a Bayesian total error analysis of conceptual rainfall-runoff models: Characterising model error using storm-dependent parameters , 2006 .

[17]  Bradley P. Carlin,et al.  Markov Chain Monte Carlo in Practice: A Roundtable Discussion , 1998 .

[18]  Witold F. Krajewski,et al.  Sampling Errors of Tipping-Bucket Rain Gauge Measurements , 2001 .

[19]  Stefania Tamea,et al.  Verification tools for probabilistic forecasts of continuous hydrological variables , 2006 .

[20]  David J. C. MacKay,et al.  A Practical Bayesian Framework for Backpropagation Networks , 1992, Neural Computation.

[21]  Faming Liang,et al.  Estimating uncertainty of streamflow simulation using Bayesian neural networks , 2009 .

[22]  Martin F. Lambert,et al.  Bayesian training of artificial neural networks used for water resources modeling , 2005 .

[23]  T. Louis,et al.  Bayes and Empirical Bayes Methods for Data Analysis. , 1997 .

[24]  Raghavan Srinivasan,et al.  Simultaneous calibration of surface flow and baseflow simulations: a revisit of the SWAT model calibration framework , 2011 .

[25]  Indrajeet Chaubey,et al.  A simplified approach to quantifying predictive and parametric uncertainty in artificial neural network hydrologic models , 2007 .

[26]  N. Metropolis,et al.  Equation of State Calculations by Fast Computing Machines , 1953, Resonance.

[27]  George Kuczera,et al.  Critical evaluation of parameter consistency and predictive uncertainty in hydrological modeling: A case study using Bayesian total error analysis , 2009 .

[28]  Xuesong Zhang,et al.  Calibration and uncertainty analysis of the SWAT model using Genetic Algorithms and Bayesian Model Averaging , 2009 .

[29]  R. L. Winkler,et al.  Scoring Rules for Continuous Probability Distributions , 1976 .

[30]  Jacob M. Montgomery,et al.  Bayesian Model Averaging: Theoretical Developments and Practical Applications , 2010, Political Analysis.

[31]  Paulin Coulibaly,et al.  Bayesian neural network for rainfall‐runoff modeling , 2006 .

[32]  Christopher M. Bishop,et al.  Neural networks for pattern recognition , 1995 .

[33]  H. Riedwyl.  Goodness of Fit , 1967 .

[34]  Raghavan Srinivasan,et al.  GIS‐Based Spatial Precipitation Estimation: A Comparison of Geostatistical Approaches 1 , 2009 .

[35]  Cajo J. F. ter Braak,et al.  Treatment of input uncertainty in hydrologic modeling: Doing hydrology backward with Markov chain Monte Carlo simulation , 2008 .

[36]  Turgay Partal,et al.  Estimation and forecasting of daily suspended sediment data using wavelet–neural networks , 2008 .

[37]  B. Bobée,et al.  Improving extreme hydrologic events forecasting using a new criterion for artificial neural network selection , 2001 .

[38]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[39]  Li-Chiu Chang,et al.  Counterpropagation fuzzy-neural network for city flood control system , 2008 .

[40]  D. Legates,et al.  Evaluating the use of “goodness‐of‐fit” Measures in hydrologic and hydroclimatic model validation , 1999 .

[41]  Raghavan Srinivasan,et al.  GIS-based spatial precipitation estimation using next generation radar and raingauge data , 2010, Environ. Model. Softw..

[42]  Roman Krzysztofowicz,et al.  The case for probabilistic forecasting in hydrology , 2001 .

[43]  Faming Liang,et al.  Bayesian neural networks for nonlinear time series forecasting , 2005, Stat. Comput..

[44]  W. Wong,et al.  Real-Parameter Evolutionary Monte Carlo With Applications to Bayesian Mixture Models , 2001 .

[45]  L. Feyen,et al.  Assessing parameter, precipitation, and predictive uncertainty in a distributed hydrological model using sequential data assimilation with the particle filter , 2009 .

[46]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[47]  Holger R. Maier,et al.  Selection of input variables for data driven models: An average shifted histogram partial mutual information estimator approach , 2009 .

[48]  Holger R. Maier,et al.  Neural networks for the prediction and forecasting of water resource variables: a review of modelling issues and applications , 2000, Environ. Model. Softw..

[49]  Alberto Montanari,et al.  Large sample behaviors of the generalized likelihood uncertainty estimation (GLUE) in assessing the uncertainty of rainfall‐runoff simulations , 2005 .

[50]  Peter Müller,et al.  Issues in Bayesian Analysis of Neural Network Models , 1998, Neural Computation.

[51]  Amin Elshorbagy,et al.  On the relevance of using artificial neural networks for estimating soil moisture content , 2008 .

[52]  Arnaud Doucet,et al.  Sequential Monte Carlo Methods to Train Neural Network Models , 2000, Neural Computation.

[53]  H. Haario,et al.  An adaptive Metropolis algorithm , 2001 .

[54]  Min Xu,et al.  Application of Bayesian regularized BP neural network model for analysis of aquatic ecological data-a case study of chlorophyll-a prediction in Nanzui water area of Dongting Lake. , 2005, Journal of environmental sciences.

[55]  J. Nash,et al.  River flow forecasting through conceptual models part I — A discussion of principles☆ , 1970 .

[56]  David B. Dunson,et al.  Bayesian Data Analysis , 2010 .

[57]  Dan Cornford,et al.  Neural Network Modelling with Input Uncertainty: Theory and Application , 2000, J. VLSI Signal Process..

[58]  S. Popescu,et al.  Bayesian Learning with Gaussian Processes for Supervised Classification of Hyperspectral Data , 2008 .

[59]  Geoffrey E. Hinton,et al.  Bayesian Learning for Neural Networks , 1995 .

[60]  Raghavendra B. Jana,et al.  Multiscale Bayesian neural networks for soil water content estimation , 2008 .

[61]  Chandranath Chatterjee,et al.  Uncertainty assessment and ensemble flood forecasting using bootstrap based artificial neural networks (BANNs) , 2010 .

[62]  Keith Beven,et al.  The future of distributed models: model calibration and uncertainty prediction. , 1992 .

[63]  Fi-John Chang,et al.  Evolutionary artificial neural networks for hydrological systems forecasting , 2009 .

[64]  Holger R. Maier,et al.  Non-linear variable selection for artificial neural networks using partial mutual information , 2008, Environ. Model. Softw..

[65]  J. Sheridan.  RAINFALL-STREAMFLOW RELATIONS FOR COASTAL PLAIN WATERSHEDS , 1997, Applied Engineering in Agriculture.

[66]  null null,et al.  Artificial Neural Networks in Hydrology. II: Hydrologic Applications , 2000 .