World water resources are under stress, and climate change, unplanned urban and rural development have generated the need for new approaches in problem-solving. Large organizations and many countries have focused efforts on the idea of data collection and open data policies. However, still many countries do lack knowledge on how to optimize the use of data. Physical and engineering formulations have been developed on the concepts of representation and limitations of data. However, nowadays, data-driven or machine learning concepts can be applied to many water resources problems.
The availability of a large amount of data (big data), including satellite products, global numerical weather models and integrated data sets with local measurements, has made it essential to be able to learn more about information extraction and use. Using Machine learning, in many data problems, has proven to be more accurate and faster than physical models. However, still more work is needed in the representation of solutions and uncertainty reduction, via the used combined physics and better analysis of the machine learning results. which allows for a better understanding. This combined information has been developing slowly, and since more than a decade machine learning problems have developed including more and more information from physics in the process of solving a problem. There are very few books dealing with the combined use of advanced machine learning and state of the art experiences in water resources. Hydroinformatics: Analysis, Modeling, and Forecasting would lay in the area of hydroinformatics and will include modeling and machine learning problems theory and examples, covering advanced computational algorithm for classification and regression problems.
Volume highlights include:
Hydroinformatics, Modelling and machine learning theory in water resources
Predicting extreme spatiotemporal hydrometeorological phenomena using machine learning
General applications of machine learning to optimal management, monitoring and use of water resources
List of Contributors vii
Preface xi
1 Hydroinformatics and Applications of Artificial Intelligence and Machine
Learning in Water-RelatedProblems 1
Gerald A. Corzo Perez and Dimitri P. Solomatine
Part I Modeling Hydrological Systems
2 Improving Model Identifiability by Driving Calibration With Stochastic
Inputs 41
Andreas Efstratiadis, Ioannis Tsoukalas, and Panagiotis Kossieris
3 A Two-Stage Surrogate-Based Parameter Calibration Framework for a Complex
DistributedHydrological Model 63
Haiting Gu, Yue-Ping Xu, Li Liu, Di Ma, Suli Pan, and Jingkai Xie
4 Fuzzy Committees of Conceptual Distributed Model 99
Mostafa Farrag, Gerald A. Corzo Perez, and Dimitri P. Solomatine
5 Regression-Based Machine Learning Approaches for Daily Streamflow Modeling
129
Vidya S. Samadi, Sadgeh Sadeghi Tabas, Catherine A. M. E. Wilson, and Daniel
R. Hitchcock
6 Use of Near-Real-Time Satellite Precipitation Data and Machine Learning to
Improve Extreme RunoffModeling 149
Paul Muñoz, Gerald A. Corzo Perez, Dimitri P. Solomatine, Jan Feyen, and
Rolando Célleri
Part II Forecasting Water Resources
7 Forecasting Water Levels Using Machine (Deep) Learning to Complement
Numerical Modeling in theSouthern Everglades, USA 179
Courtney S. Forde, Biswa Bhattacharya, Dimitri P. Solomatine, Eric D. Swain,
and Nicholas G. Aumen
8 Application of a Multilayer Perceptron Artificial Neural Network (MLP-ANN)
in HydrologicalForecasting in El Salvador 213
Jose Valles
9 Noise Filter With Wavelet Analysis in Artificial Neural Networks (NOWANN)
for Flow Time SeriesPrediction 241
Daniel A. Vázquez, Gerald A. Corzo Perez, and Dimitri P. Solomatine
Part III Knowledge Discovery and Optimization
10 Application of Natural Language Processing to Identify Extreme
Hydrometeorological Events inDigital News Media: Case of the Magdalena River
Basin, Colombia 285
Santiago Duarte, Gerald A. Corzo Perez, Germán Santos, and Dimitri P.
Solomatine
11 Three-Dimensional Clustering in the Characterization of Spatiotemporal
Drought Dynamics: ClusterSize Filter and Drought Indicator Threshold
Optimization 319
Vitali Diaz, Gerald A. Corzo Perez, Henny A. J. Van Lanen, and Dimitri P.
Solomatine
12 Deep Learning of Extreme Rainfall Patterns Using Enhanced Spatial Random
Sampling With PatternRecognition 343
Han Wang and Yunqing Xuan
13 Teleconnection Patterns of River Water Quality Dynamics Based on Complex
Network Analysis 357
Jiping Jiang, Sijie Tang, Bellie Sivakumar, Tianrui Pang, Na Wu, and Yi
Zheng
14 Probabilistic Analysis of Flood Storage Areas Management in the Huai
River Basin, China, WithRobust Optimization and Similarity-Based Selection
for Real-Time Operation 373
Xingyu Zhou, Andreja Jonoski, Ioana Popescu, and Dimitri P. Solomatine
15 Multi-Objective Optimization of Reservoir Operation Policies Using
Machine Learning Models: ACase Study of the Hatillo Reservoir in the
Dominican Republic 409
Carlos Tami, Gerald A. Corzo Perez, Fidel Perez, and Germain Santos
Index 447
Gerald A. Corzo Perez, IHE Delft Institute for Water Education, The Netherlands
Dimitri P. Solomatine, IHE Delft Institute for Water Education, and Delft University of Technology, The Netherlands, and Water Problems Institute of the Russian Academy of Sciences, Moscow, Russia
