H112
Unsupervised learning in Hydrology: Harnessing Big Data to advance process understanding
Session ID#: 158261
Session Description:
The ever-growing body of data from remote sensing, in-situ networks, and Earth system models has led to exabytes of valuable yet underutilized datasets; this is a missed opportunity to advance process understanding in hydrology. The underuse of these data has primarily been driven by a lack of approaches to effectively and efficiently synthesize understanding across large and diverse datasets. There are many innovative, scalable methods for unsupervised learning – such as novel approaches to clustering and dimensionality reduction, or learning graph representations using graph neural networks, among others – which provide unique opportunities to find deeper understanding within these data. This session aims to bring together hydrologists and data scientists that are using unsupervised learning techniques to leverage available datasets to advance process understanding in hydrology across all spatial and temporal scales.
Index Terms:
0555 Neural networks, fuzzy logic, machine learning [COMPUTATIONAL GEOPHYSICS]
1805 Computational hydrology [HYDROLOGY]
1819 Geographic Information Systems (GIS) [HYDROLOGY]
1836 Hydrological cycles and budgets [HYDROLOGY]
Primary Convener: Nathaniel W. Chaney, Duke University, Civil and Environmental Engineering, Durham, NC, United States
Conveners: Haruko M Wainwright, Lawrence Berkeley National Laboratory, Berkeley, CA, United States, Richard T Mills, Oak Ridge National Lab, Oak Ridge, TN, United States and Nilay Dogulu, Independent Researcher, Ankara, Turkey
Student/Early Career Convener: Noemi Vergopolan, Princeton / GFDL, Atmospheric and Ocean Sciences Program, Princeton, NJ, United States
Flood Classification Based on Hydrograph Characteristics (1128497)
Abinesh Ganapathy1, David M Hannah2 and Ankit Agarwal1, (1)Indian Institute of Technology Roorkee, Department of Hydrology, Roorkee, India, (2)University of Birmingham, School of Geography, Earth and Environmental Sciences, Birmingham, United Kingdom
Interpretable Classification of the Contiguous United States River Catchments using Network Science Methods (1132661)
Fabio Ciulla1, Jared Willard2, Helen Weierbach3 and Charuleka Varadharajan1, (1)Lawrence Berkeley National Laboratory, Berkeley, CA, United States, (2)University of Minnesota, Minneapolis, MN, United States, (3)Lawrence Berkeley National Laboratory, Earth and Environmental Sciences Area, Berkeley, United States
Using Unsupervised Learning to Determine Dominant Physical Features that Control the Land Surface Temperature Variability over the CONUS Mountain West (1156988)
Laura Torres Torres Rojas1, Haruko M Wainwright2,3, Yuxin Wu4 and Nathaniel W. Chaney1, (1)Duke University, Civil and Environmental Engineering, Durham, NC, United States, (2)Lawrence Berkeley National Laboratory, Berkeley, CA, United States, (3)Massachusetts Institute of Technology, Cambridge, United States, (4)Lawrence Berkeley National Laboratory, Earth and Environmental Sciences Area, Berkeley, CA, United States
Surface-Groundwater Interaction in a Tropical Catchment: Estimating Endmember δ2H & δ18O Isotopes Using Deep Learning Models (1164378)
Siva Naga Venkat Venkat Nara, Indian Institute of Science, Interdisciplinary Centre for Water Research, Bangalore, India, Prosenjit Ghosh, Indian Institute of Science, Centre for Earth Sciences, Divecha Centre for Climate Change, Interdisciplinary Centre for Water Research, Bangalore, India and Sekhar Muddu, Indian Institute of Science, Department of Civil Engineering, Interdisciplinary Centre for Water Research, Bangalore, India
Cluster Analysis and Multiclass Prediction to Discover Flood Generation Mechanisms Across the Conterminous United States (1165250)
Ken Eng, U.S. Geological Survey, Reston, VA, United States, Jared David Smith, U.S. Geological Survey, Integrated Modeling and Prediction Division, Reston, VA, United States, Sara Levin, USGS Upper Midwest Environmental Sciences Center, La Crosse, United States, Charles Stillwell, U.S. Geological Survey, Raleigh, United States and Andrew Sekellick, USGS Maryland/Delaware/District of Columbia Water Science Center, Baltimore, MD, United States
Unsupervised Machine Learning Methods for Predicting Droughts (1180354)
Vishwas Rao1, Amit Subrahmanya2, Julie Bessac3, Brandi Gamelin4 and Mustafa Altinakar3, (1)Argonne National Laboratory, Argonne, IL, United States, (2)Virginia Polytechnic Institute and State University, Blacksburg, United States, (3)Argonne National Laboratory, Argonne, United States, (4)California State University Los Angeles, Chino Hills, CA, United States
