A Bayesian Neural Network for an Accurate Representation and Transformation of Runoff Dynamics: A Case Study of the Brazos River Basin in Texas
DOI:
https://doi.org/10.12974/2311-8741.2020.08.5Keywords:
Variational Bayesian neural network, VIC Model, Similarity, Transfer-Learning, Pearson correlation coefficient.Abstract
Conventional physically based models have long yielded promising results, as they have been the main tool to depict the underpinnings of the physics governing the hydrological events. These models, however, suffer from certain issues such as the intense calibration time or the uncertainty in the estimation of hydrological variables. The development of the sophisticated data-driven techniques, and machine learning models in particular, combined with rapid increases in computational abilities (graphics processing units, computer clusters. etc.), has enabled hydrologists to utilize the data driven models in tandem with the well-established hydrological models to simulate miscellaneous environmental processes nimbly, and therefore circumvent the aforementioned conundrums associated with the physically based models. To this end, the present study aims at exploring a sophisticated neural network called variational Bayesian neural network, to improve the accuracy of physically based predictions such as runoff. Our neural network was able to accurately forecast the runoff rates with the mean Pearson correlation coefficient of 86.27%+0.0599 within a randomly selected subset of cells in the Brazos River Basin. As these cells are selected randomly across the basin, we exclude the possibility of biasing our neural network by any specific cell. Moreover, this work for the very first time, to the best of our knowledge, suggests a similarity-based solution to transfer the learning model developed in a basin to be deployed across a different basin. In other words, there would be no need to develop a learning model for each basin from scratch. We, instead, utilize the models learnt from the previously studied basins. We cross-validated our proposed transfer learning solution via leave-one-out strategy within the grid cells of the Brazos River basin achieving a mean Pearson correlation coefficient of 85.83%+0.0592.
References
Hu C, Wu Q, Li H, Jian S, Li N, Lou Z. Deep learning with a long short-term memory networks approach for rainfall-runoff simulation. Water. 2018 Oct 30; 10(11): 1543. https://doi.org/10.3390/w10111543
Sorooshian S, Gupta VK. Automatic calibration of conceptual rainfall‐runoff models: The question of parameter observability and uniqueness. Water Resources Research. 1983 Feb; 19(1): 260-8. https://doi.org/10.1029/WR019i001p00260
Beven K, Freer J. Equifinality, data assimilation, and uncertainty estimation in mechanistic modelling of complex environmental systems using the GLUE methodology. Journal of hydrology. 2001 Aug 1; 249(1-4): 11-29. https://doi.org/10.1016/S0022-1694(01)00421-8
Madsen H. Parameter estimation in distributed hydrological catchment modelling using automatic calibration with multiple objectives. Advances in water resources. 2003 Feb 1; 26(2): 205-16. https://doi.org/10.1016/S0309-1708(02)00092-1
Geetha A, Nasira GM. Time-series modelling and forecasting: Modelling of rainfall prediction using ARIMA model. International Journal of Society Systems Science. 2016; 8(4): 361-72. https://doi.org/10.1504/IJSSS.2016.081411
Graham A, Mishra EP. Time series analysis model to forecast rainfall for Allahabad region. Journal of Pharmacognosy and Phytochemistry. 2017; 6(5): 1418-21.
Stern RD, Coe R. A model fitting analysis of daily rainfall data. Journal of the Royal Statistical Society: Series A (General). 1984 Jan; 147(1): 1-8. https://doi.org/10.2307/2981736
Chandler RE, Wheater HS. Analysis of rainfall variability using generalized linear models: a case study from the west of Ireland. Water Resources Research 2002 Oct; 38(10): 10-. https://doi.org/10.1029/2001WR000906
Stampoulis D, Reager JT, David CH, Andreadis KM, Famiglietti JS, Farr TG, Trangsrud AR, Basilio RR, Sabo JL, Osterman GB, Lundgren PR. Model-data fusion of hydrologic simulations and GRACE terrestrial water storage observations to estimate changes in water table depth. Advances in Water Resources. 2019 Jun 1; 128: 13-27. https://doi.org/10.1016/j.advwatres.2019.04.004
Girotto M, De Lannoy GJ, Reichle RH, Rodell M. Assimilation of gridded terrestrial water storage observations from GRACE into a land surface model. Water Resources Research. 2016 May; 52(5): 4164-83. https://doi.org/10.1002/2015WR018417
Girotto M, De Lannoy GJ, Reichle RH, Rodell M, Draper C, Bhanja SN, Mukherjee A. Benefits and pitfalls of GRACE data assimilation: A case study of terrestrial water storage depletion in India. Geophysical research letters. 2017 May 16; 44(9): 4107-15. https://doi.org/10.1002/2017GL072994
Su H, Yang ZL, Dickinson RE, Wilson CR, Niu GY. Multisensor snow data assimilation at the continental scale: The value of Gravity Recovery and Climate Experiment terrestrial water storage information. Journal of Geophysical Research: Atmospheres. 2010 May 27; 115(D10). https://doi.org/10.1029/2009JD013035
Houborg R, Rodell M, Li B, Reichle R, Zaitchik BF. Drought indicators based on model‐assimilated Gravity Recovery and Climate Experiment (GRACE) terrestrial water storage observations. Water Resources Research. 2012 Jul; 48(7). https://doi.org/10.1029/2011WR011291
Forman BA, Reichle RH, Rodell M. Assimilation of terrestrial water storage from GRACE in a snow‐dominated basin. Water Resources Research. 2012 Jan; 48(1). https://doi.org/10.1029/2011WR011239
Reager J, Thomas A, Sproles E, Rodell M, Beaudoing H, Li B, Famiglietti J. Assimilation of GRACE terrestrial water storage observations into a land surface model for the assessment of regional flood potential. Remote Sensing. 2015 Nov; 7(11): 14663-79. https://doi.org/10.3390/rs71114663
Tripathi S, Srinivas VV, Nanjundiah RS. Downscaling of precipitation for climate change scenarios: a support vector machine approach. Journal of hydrology. 2006 Nov 15; 330(3-4): 621-40. https://doi.org/10.1016/j.jhydrol.2006.04.030
Shrestha DL, Solomatine DP. Data‐driven approaches for estimating uncertainty in rainfall‐runoff modelling. International Journal of River Basin Management. 2008 Jun 1; 6(2): 109-22. https://doi.org/10.1080/15715124.2008.9635341
Selle B, Muttil N. Testing the structure of a hydrological model using Genetic Programming. Journal of Hydrology. 2011 Jan 24; 397(1-2): 1-9. https://doi.org/10.1016/j.jhydrol.2010.11.009
Kratzert F, Klotz D, Brenner C, Schulz K, Herrnegger M. Rainfall-runoff modelling using long short-term memory (LSTM) networks. Hydrology and Earth System Sciences. 2018 Nov 22; 22(11): 6005-22. https://doi.org/10.5194/hess-22-6005-2018
Alizadeh Z, Yazdi J, Kim J, Al-Shamiri A. Assessment of Machine Learning Techniques for Monthly Flow Prediction. Water. 2018 Nov 17; 10(11): 1676. https://doi.org/10.3390/w10111676
Kenabatho PK, Parida BP, Moalafhi DB, Segosebe T. Analysis of rainfall and large-scale predictors using a stochastic model and artificial neural network for hydrological applications in southern Africa. Hydrological Sciences Journal. 2015 Nov 2; 60(11): 1943-55.
Fang K, Shen C, Kifer D, Yang X. Prolongation of SMAP to spatiotemporally seamless coverage of continental US using a deep learning neural network. Geophysical Research Letters. 2017 Nov 16; 44(21): 11-030. https://doi.org/10.1002/2017GL075619
Tokar AS, Johnson PA. Rainfall-runoff modeling using artificial neural networks. Journal of Hydrologic Engineering. 1999 Jul; 4(3): 232-9. https://doi.org/10.1061/(ASCE)1084-0699(1999)4:3(232)
Vehtari A, Gelman A, Gabry J. Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. Statistics and computing. 2017 Sep 1; 27(5): 1413-32. https://doi.org/10.1007/s11222-016-9696-4
Behrangi A, Yin X, Rajagopal S, Stampoulis D, Ye H. On distinguishing snowfall from rainfall using near‐surface atmospheric information: C omparative analysis, uncertainties and hydrologic importance. Quarterly Journal of the Royal Meteorological Society. 2018 Nov; 144: 89-102. https://doi.org/10.1002/qj.3240
Tayfur G, Singh VP. ANN and fuzzy logic models for simulating event-based rainfall-runoff. Journal of hydraulic engineering. 2006 Dec; 132(12): 1321-30. https://doi.org/10.1061/(ASCE)0733- 9429(2006)132:12(1321)
Dawson CW, Wilby R. An artificial neural network approach to rainfall-runoff modelling. Hydrological Sciences Journal. 1998 Feb 1; 43(1): 47-66. https://doi.org/10.1080/02626669809492102
Maurer EP, Wood AW, Adam JC, Lettenmaier DP, Nijssen B. A long-term hydrologically based dataset of land surface fluxes and states for the conterminous United States. Journal of climate. 2002 Nov; 15(22): 3237-51. https://doi.org/10.1175/1520- 0442(2002)015<3237:ALTHBD>2.0.CO;2
Livneh B, Rosenberg EA, Lin C, Nijssen B, Mishra V, Andreadis KM, Maurer EP, Lettenmaier DP. A long-term hydrologically based dataset of land surface fluxes and states for the conterminous United States: Update and extensions. Journal of Climate. 2013 Dec; 26(23): 9384-92. https://doi.org/10.1175/JCLI-D-12-00508.1
Blundell C, Cornebise J, Kavukcuoglu K, Wierstra D. Weight uncertainty in neural networks. arXiv preprint arXiv: 1505.05424. 2015 May 20.
Durrett R. Probability: theory and examples. Cambridge university press; 2019 Apr 18. https://doi.org/10.1017/9781108591034
Myung IJ. Tutorial on maximum likelihood estimation. Journal of mathematical Psychology. 2003 Feb 1; 47(1): 90-100. https://doi.org/10.1016/S0022-2496(02)00028-7
Korb KB, Nicholson AE. Bayesian artificial intelligence. CRC press; 2010 Dec 16. https://doi.org/10.1201/b10391
Gauvain JL, Lee CH. Maximum a posteriori estimation for multivariate Gaussian mixture observations of Markov chains. IEEE transactions on speech and audio processing. 1994 Apr; 2(2): 291-8. https://doi.org/10.1109/89.279278
Scholkopf B, Smola AJ. Learning with kernels: support vector machines, regularization, optimization, and beyond. MIT press; 2001 Dec 1.
Dietterich T. Overfitting and undercomputing in machine learning. ACM computing surveys. 1995 Sep 1; 27(3): 326-7. https://doi.org/10.1145/212094.212114
Hershey JR, Olsen PA. Approximating the Kullback Leibler divergence between Gaussian mixture models. In2007 IEEE International Conference on Acoustics, Speech and Signal Processing-ICASSP'07 2007 Apr 15 (Vol. 4, pp. IV-317). IEEE. https://doi.org/10.1109/ICASSP.2007.366913
Zhang Z. Improved Adam Optimizer for Deep Neural Networks. In2018 IEEE/ACM 26th International Symposium on Quality of Service (IWQoS) 2018 Jun 4 (pp. 1-2). IEEE. https://doi.org/10.1109/IWQoS.2018.8624183
https: //www.coursera.org/lecture/machine-learning/anomalydetection- using-the-multivariate-gaussian-distribution-DnNr9
Dai W, Jin O, Xue GR, Yang Q, Yu Y. Eigentransfer: a unified framework for transfer learning. InProceedings of the 26th Annual International Conference on Machine Learning 2009 Jun 14 (pp. 193-200). ACM. https://doi.org/10.1145/1553374.1553399
Pan SJ, Yang Q. A survey on transfer learning. IEEE Transaction on Knowledge Discovery and Data Engineering, 22 (10). https://doi.org/10.1109/TKDE.2009.191
Salvador S, Chan P. Toward accurate dynamic time warping in linear time and space. Intelligent Data Analysis. 2007 Jan 1; 11(5): 561-80. https://doi.org/10.3233/IDA-2007-11508
Damavandi HG, Gupta AS, Nelson RK, Reddy CM. Interpreting comprehensive two-dimensional gas chromatography using peak topography maps with application to petroleum forensics. Chemistry Central Journal. 2016 Dec; 10(1): 75. https://doi.org/10.1186/s13065-016-0211-y
Damavandi HG, Gupta AS, Reddy C, Nelson R. Oil-spill forensics using two-dimensional gas chromatography: Differentiating highly correlated petroleum sources using peak manifold clusters. In2015 49th Asilomar Conference on Signals, Systems and Computers 2015 Nov 8 (pp. 1589- 1592). IEEE. https://doi.org/10.1109/ACSSC.2015.7421415
Sumi SM, Zaman MF, Hirose H. A rainfall forecasting method using machine learning models and its application to the Fukuoka city case. International Journal of Applied Mathematics and Computer Science. 2012 Dec 1; 22(4): 841-54. https://doi.org/10.2478/v10006-012-0062-1
Kim TW, Valdés JB, Aparicio J. Frequency and spatial characteristics of droughts in the Conchos River Basin, Mexico. Water International. 2002 Sep 1; 27(3): 420-30. https://doi.org/10.1080/02508060208687021
Karran DJ, Morin E, Adamowski J. Multi-step streamflow forecasting using data-driven non-linear methods in contrasting climate regimes. Journal of Hydroinformatics. 2013 Nov 6; 16(3): 671-89. https://doi.org/10.2166/hydro.2013.042
Khan MS, Coulibaly P. Application of support vector machine in lake water level prediction. Journal of Hydrologic Engineering. 2006 May; 11(3): 199-205. https://doi.org/10.1061/(ASCE)1084-0699(2006)11:3(199)
Damavandi HG, Shah R, Stampoulis D, Wei Y, Boscovic D, Sabo J. Accurate prediction of streamflow using long shortterm memory network: A case study in the Brazos river basin in Texas. International Journal of Environmental Science and Development. 2019 Jan 1; 10(10): 294-300. https://doi.org/10.18178/ijesd.2019.10.10.1190
Damavandi HG, Stampoulis D, Shah R, Wei Y, Boscovic D, Sabo J. Machine learning: An efficient alternative to the variable infiltration capacity model for an accurate simulation of runoff rates. International Journal of Environmental Science and Development. 2019 Jan 1; 10(9): 288-93. https://doi.org/10.18178/ijesd.2019.10.9.1189