H093
Machine Learning in Hydrologic Modeling
Session ID#: 75246
Session Description:
Machine learning (ML) is now standard across the Earth Sciences. The goal of this session is to highlight novel or practically-valuable applications of machine learning in Hydrologic modeling.
In their summary of a workshop on 'Big Data and the Earth Sciences', Sellars (2018) noted that “participants who worked in modeling physical-based systems continue to raise caution about the lack of physical understanding of machine learning methods." A critical challenge is to develop strategies to utilize strengths and mitigate weaknesses in each/both types of modeling approaches. This session solicits abstracts related to integrating ML into traditional hydrological modeling tasks and workflows. This might include (1) improved understanding of under what conditions ML provides reliable vs. unreliable simulations, (2) imposing physical constraints into ML models, (3) integrating ML into process-based models or workflows such as parameter estimation or inverse modeling, or (4) deriving scaling relationships or process-related insights directly from ML.
Index Terms:
1847 Modeling [HYDROLOGY]
1873 Uncertainty assessment [HYDROLOGY]
1906 Computational models, algorithms [INFORMATICS]
1942 Machine learning [INFORMATICS]
Primary Convener: Grey Stephen Nearing, University of Alabama, Geological Sciences, Tuscaloosa, AL, United States
Conveners: Hoshin Vijai Gupta, University of Arizona, Department of Hydrology and Atmospheric Sciences, Tucson, AZ, United States, Chaopeng Shen, Pennsylvania State University Main Campus, Civil and Environmental Engineering, University Park, PA, United States and Xingyuan Chen, Pacific Northwest National Laboratory, Richland, WA, United States
Primary Liaison: Grey Stephen Nearing, NASA Goddard Space Flight Center, Greenbelt, MD, United States
Physically-Based Machine Learning for Hydrological Modeling (491651)
Grey Stephen Nearing1, Hoshin Vijai Gupta2, Frederik Kratzert3, Daniel Klotz3 and Alden Keefe Sampson4, (1)University of Alabama, Geological Sciences, Tuscaloosa, AL, United States, (2)University of Arizona, Department of Hydrology and Atmospheric Sciences, Tucson, AZ, United States, (3)Johannes Kepler University, Institute for Machine Learning, Linz, Austria, (4)Upstream Tech, Alameda, CA, United States
Integrating system modelling and machine learning for (eco-)hydrological research (Invited) (492917)
Markus Reichstein1, Pierre Gentine2, Basil Kraft3, Martin Jung4, Sungmin O5, Rene Orth4 and Wenli Zhao6, (1)Max Planck Institute for Biogeochemistry, Department of Biogeochemical Integration, Jena, Germany, (2)Department of Earth and Environmental Engineering, Columbia University, New York, NY, United States, (3)Max-Planck-Institute for Biogeochemistry, Department of Biogeochemical Integration, Jena, Germany, (4)Max Planck Institute for Biogeochemistry, Jena, Germany, (5)University of Graz, Institute for Geophysics, Astrophysics, and Meteorology/Institute of Physics (IGAM/IP), NAWI, Graz, Austria, (6)Peking University Shenzhen Graduate School, Shenzhen, China
Function Space Optimization (FSO): A novel method for estimating parameter transfer functions for hydrological models (493146)
Moritz Feigl1, Mathew Herrnegger1, Daniel Klotz2 and Karsten Schulz1, (1)BOKU University of Natural Resources and Life Sciences, Institute for Hydrology and Water Management, Vienna, Austria, (2)Johannes Kepler University, Institute for Machine Learning, Linz, Austria
Physics-constrained machine learning of evapotranspiration (500357)
Wenli Zhao1,2, Pierre Gentine3, Markus Reichstein4, Yao Zhang5, Yeqiang Wen6, Xi Li7, Changjie Lin6, Sha Zhou2 and Guo Yu Qiu1, (1)Peking University Shenzhen Graduate School, Shenzhen, China, (2)Columbia University, Earth and Environmental Engineering Department, New York, NY, United States, (3)Department of Earth and Environmental Engineering, Columbia University, New York, NY, United States, (4)Max Planck Institute for Biogeochemistry, Department of Biogeochemical Integration, Jena, Germany, (5)Department of Earth and Environmental Engineering, Columbia University, New York, United States, (6)Tsinghua University, Beijing, China, (7)Peking University, Beijing, China
Estimating location of suitable drinking groundwater from GIS-based data using tree-based ensemble machine learning (501014)
Hye Lim Lee1, Dugin Kaown2, Eun-Hee Koh1, Sanghoon Lee3 and Kang Kun Lee1, (1)Seoul National University, Seoul, South Korea, (2)Seoul Nation Univ, Seoul, Korea, Republic of (South), (3)Seoul National University, Seoul, Korea, Republic of (South)
Evaluation and machine learning improvement of global flood simulations (503041)
Tao Yang1,2, Fubao Sun1, Pierre Gentine3,4, Wenbin Liu1, Hong Wang1, Jiabo Yin2,5, Muye Du1 and Liu Changming1, (1)Key Laboratory of Water Cycle and Related Land Surface Processes, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, China, (2)Department of Earth and Environmental Engineering, Columbia University, New York, United States, (3)Earth Institute, Columbia University, New York, NY, United States, (4)Department of Earth and Environmental Engineering, Columbia University, New York, NY, United States, (5)State Key Laboratory of Water Resources and Hydropower Engineering Science, Wuhan University, Wuhan, China
Predicting daily streamflow with a dilated convolutional neural network. (504084)
Shiheng Duan, Univ California Davis, Davis, CA, United States, Paul Aaron Ullrich, Univ California Davis, Fremont, CA, United States, Lele Shu, Pennsylvania State University, University Park, PA, United States and David Hall, NVIDIA, Lafayette, CO, United States
Evaluating impacts of input variables in deep neural network model for finding locations of ‘good’ groundwater (504669)
Sanghoon Lee1, Dugin Kaown1, Eun-Hee Koh2, Hye Lim Lee2 and Kang Kun Lee2, (1)Seoul National University, Seoul, Korea, Republic of (South), (2)Seoul National University, Seoul, South Korea
Learning from the data: manifold learning in interpreting tracers of the landscape hydrologic system (506539)
Fei Xu, Johns Hopkins University, Baltimore, MD, United States, Ciaran J Harman, Johns Hopkins University, Environmental Health and Engineering, Baltimore, MD, United States, Mauro Maggioni, Johns Hopkins University, Applied Mathematics and Statistics, Baltimore, United States and Fei Lu, Johns Hopkins University, Mathematics, Baltimore, United States
Modeling and Forecasting Human Modified Streamflow Using a Recurrent Neural Network (507051)
Tushar Khurana1,2, Eliza Hale3, Alden Keefe Sampson1 and David Lambl3, (1)Upstream Tech, Alameda, CA, United States, (2)University of Washington Seattle, Seattle, WA, United States, (3)Natel Energy Inc, Upstream Tech, Alameda, CA, United States
Application of Machine-Learning Techniques to Pareto-Optimal Groundwater Management Solutions (508259)
Tracy Nishikawa1, Zachary P Stanko1 and Brian J Wagner2, (1)USGS California Water Science Center San Diego, San Diego, CA, United States, (2)USGS Geological Survey, Earth System Processes Division, Water Cycle Branch, Menlo Park, CA, United States
Imagining and Building the Next Generation of the US Department of Agriculture Natural Resources Conservation Service Operational Water Supply Forecast System for the American West: Machine Learning Goes Mainstream (508801)
Sean William Fleming, White Rabbit R&D LLC, Corvallis, OR, United States, Angus Graham Goodbody, Natural Resource Conservation Service, Portland, United States, David C Garen, NRCS, Portland, OR, United States and Cara McCarthy, Natural Resources Conservation Service, Portland, United States
Quantifying landscape-water quality nexus and predicting water quality under alternative planning scenarios using machine learning approaches (509992)
Runzi Wang, Michigan State University, East Lansing, MI, United States, Liangji Kong, Peking University, Beijing, China and Ming-Han Li, Michigan State University, East Lansing, United States
Parameter estimation on dynamic manifolds (511912)
Francesca Boso, Stanford University, Stanford, CA, United States and Daniel M Tartakovsky, Stanford University, Energy Resources Engineering, Stanford, CA, United States
Machine Learning for Streamflow Prediction: Current Status and Future Prospects (516700)
Martin Gauch1, Raphael Tang1, Juliane Mai2, Bryan Tolson2, Shervan Gharari3 and Jimmy Lin1, (1)University of Waterloo, David R. Cheriton School of Computer Science, Waterloo, ON, Canada, (2)University of Waterloo, Department of Civil and Environmental Engineering, Waterloo, ON, Canada, (3)University of Saskatchewan, School of Environment and Sustainability (SENS), Saskatoon, SK, Canada
Modelling the Transient Effects of Groundwater Pumping on Groundwater Storage and Surface Water using Artificial Neural Networks (520325)
Thomas Boerman1, Thomas P Gleeson2, Qiang Li1 and Samuel C Zipper3, (1)University of Victoria, Victoria, BC, Canada, (2)Civil program at UVic, Victoria, BC, Canada, (3)University of Wisconsin Madison, Madison, WI, United States
Exploring ML Tools to Identify Subsurface Voids Using Air Pressure Monitoring - Training on Coupled Unsaturated Water Flow and Air Flow Models (541561)
Ty P.A. Ferre1, Doug Denu2, Stephen Farrington3,4, Jeff Klakovich4, Andrea Pearce4 and Neal E Van Wyck4,5, (1)University of Arizona, Department of Hydrology and Atmospheric Sciences, Tucson, AZ, United States, (2)Transcend Engineering, Bethel, United States, (3)Transcend Engineering and Technology LLC, Gaysville, VT, United States, (4)Transcend Engineering, Bethel, VT, United States, (5)Organization Not Listed, Gaysville, VT, United States
Machine Learning Applied to Residence Times of Hydrological Exchange Flows in the Columbia River (548525)
Huiying Ren1, Zhangshuan Hou2, Xuehang Song1, Yilin Fang3, William A Perkins1, Huifen Zhou4 and Timothy D Scheibe1, (1)Pacific Northwest National Laboratory, Richland, WA, United States, (2)Pacific Northwest National Laboratory, Energy and Environment Directorate, Richland, WA, United States, (3)Battelle, Pacific Northwest National Laboratory, Richland, WA, United States, (4)Pacific Northwest National Laboratory, Richland, United States
Seasonal Climate Forecasting and Insights Using Teleconnections and Long Short Term Memory Networks (549261)
Charles L Becker1, Kendra Elena Kaiser1, Alejandro N Flores1, David E Wilkins1 and Nancy F Glenn2, (1)Boise State University, Boise, ID, United States, (2)Boise State Univ, Boise, ID, United States
An Ensemble of Deep Learning Models for Soil Moisture Prediction Across the Contiguous U.S. (550377)
Kathy Breen1, Scott C James2, Joseph D White3, Peter M Allen1 and Jeffrey G Arnold4, (1)Baylor University, Waco, TX, United States, (2)Baylor University, Geosciences and Mechanical Engineering, Waco, TX, United States, (3)Baylor Univ, Waco, TX, United States, (4)Texas A&M University College Station, College Station, TX, United States
Hybrid physically-based and deep learning modeling of a snow dominated mountainous karst watershed (570583)
Tianfang Xu1,2, Conor Tyson1, Qianqiu Longyang1, Ruijie Zeng1, Bethany T Neilson1 and David G Tarboton1, (1)Utah State University, Logan, UT, United States, (2)Arizona State University, Tempe, United States
Building AI-based prediction models of storm sewer systems to estimate the flooded area of the ground surface (594440)
Gary Wee1, Chia-Yu Hsu2, Kai-Yao Chang3, Kuo-Cheng Chen3, Fi-John Chang4 and Li-Chiu Chang2, (1)Tamkang University, Department of Water Resources and Environmental Engineering, New Taipei, Taiwan, (2)Tamkang University, Department of Water Resources and Environmental Engineering, Taipei, Taiwan, (3)Taipei City Government, Hydraulic Engineering Office, Public Works Department, Taipei, Taiwan, (4)National Taiwan University, Department of Bioenvironmental Systems Engineering, Taipei, Taiwan
Convolutional Neural Networks for Analyzing the Characteristics of Typhoon Structure from Satellite Imagery (595184)
Chia-Yu Hsu1, Shun-Nien Yang2, Gary Wee2 and Li-Chiu Chang1, (1)Tamkang University, Department of Water Resources and Environmental Engineering, Taipei, Taiwan, (2)Tamkang University, Department of Water Resources and Environmental Engineering, New Taipei, Taiwan
Convolutional Neural Network for Subsurface Flow Prediction (597011)
Junjie Yu1, Atefeh Jahandideh2, Syamil Mohd Razak3 and Behnam Jafarpour3, (1)Los Angeles, California, United States, (2)University of Southern California, Los Angeles, United States, (3)University of Southern California, Los Angeles, CA, United States
Applying deep learning in estimating parameters for hydrologic model (609941)
Wen-Ping Tsai1, Chaopeng Shen2 and Kuai Fang2, (1)Pennsylvania State University Main Campus, University Park, PA, United States, (2)Pennsylvania State University Main Campus, Civil and Environmental Engineering, University Park, PA, United States
A Machine Learning Approach for Data Assimilation into Terrestrial Hydrology Models (619473)
Craig Pelissier, NASA Goddard Space Flight Center, Computational and Information Science & Technology Office, Greenbelt, MD, United States, Grey Stephen Nearing, NASA Goddard Space Flight Center, Greenbelt, MD, United States and Jonathan M Frame, University of Alabama, Geological Sciences, Tuscaloosa, AL, United States
Mapping the reliability of semi-seasonal forecasts of groundwater-levels using machine learning across the Northern High Plains (622032)
Francisco Munoz-Arriola, University of Nebraska Lincoln, Lincoln, NE, United States, Alessandro Amaranto, Greifswald, Germany, Parisa Sarzaeim, University of Nebraska-Lincoln, Department of Biological Systems Engineering, Lincoln, NE, United States and Luciano Alves-Oliveira, University of Nebraska-Lincoln, Department of Biological Systems Engineering, Lincoln, United States
Develop AI Early flood warning systems for smart cities flood defense with IoT flood sensors (624720)
Shun-Nien Yang1, Ying-Yu Ku1, Chun-Ling Kuo2, Yi-Fung Wang2, Cheng-Yi Lin2, Fi-John Chang3 and Li-Chiu Chang4, (1)Tamkang University, Department of Water Resources and Environmental Engineering, New Taipei, Taiwan, (2)Water Resources Agency, Taipei, Taiwan, (3)National Taiwan University, Department of Bioenvironmental Systems Engineering, Taipei, Taiwan, (4)Tamkang University, Department of Water Resources and Environmental Engineering, Taipei, Taiwan
Applications of Deep Learning and Domain Adaptation to the Study of Mountain Snow (626666)
Joel Agustin Gongora1,2, Hans-Peter Marshall3, Christopher Hase2, Trent Bradberry2 and Samuel Ballerini2, (1)Boise State University, Geosciences, Boise, ID, United States, (2)Elder Research Inc., Arlington, VA, United States, (3)Boise State University, Department of Geosciences, Boise, ID, United States
Industry and Academic Collaboration to Advance Adoption of Machine Learning Hydrology (Invited) (627993)
Alden Keefe Sampson1, Grey Stephen Nearing2, Frederik Kratzert3 and Marshall Moutenot1, (1)Natel Energy Inc, Upstream Tech, Alameda, CA, United States, (2)University of Alabama, Geological Sciences, Tuscaloosa, AL, United States, (3)Johannes Kepler University, Institute for Machine Learning, Linz, Austria
