H014
Advances in Machine Learning for Earth Science: Observation, Modeling, and Applications
Session ID#: 249942
Session Description:
Machine learning (ML) methods have shown tremendous potential and advances in the understanding of Earth science. New datasets and ML models with high spatial and temporal resolutions are emerging at an unprecedented rate, which has opened up various new avenues of research in the field. This session aims to engage diverse earth scientific communities to share their novel ML methods and applications on radar & satellite observations, data fusion, Earth system modeling & forecast, natural hazard and extreme events, climate projection, environmental sustainability, explainable AI, etc. Studies within but not limited to hydrology and precipitation, with generalizable results are strongly encouraged.
Index Terms:
1655 Water cycles [GLOBAL CHANGE]
1854 Precipitation [HYDROLOGY]
1855 Remote sensing [HYDROLOGY]
1942 Machine learning [INFORMATICS]
Primary Convener: Siyu Zhu, University of Oklahoma, School of Civil Engineering and Environmental Science, Norman, United States
Conveners: Yixin Wen, University of Florida, Department of Geography, Ft Walton Beach, FL, United States, Guoqiang Tang, Wuhan University, School of Water Resources and Hydropower Engineering, Wuhan, China and Phu Nguyen, University of California, Irvine, Department of Civil and Environmental Engineering, Irvine, United States
Student/Early Career Convener: Mengye Chen, The University of Oklahoma, Center for Analysis and Prediction of Storms, Norman, United States
16-day Cloudless and Shadowless Composite of Harmonized Landsat Sentinel (HLS) Data for Red, NIR, SWIR1, and SWIR2 Bands (1898297)
Zhen Liu, University of Maryland College Park, College Park, MD, United States and Matthew Hansen, University of Maryland, Department of Geographical Sciences, College Park, United States
A Pretrained Foundation AI Model Reveals Global Landscape Dynamics and Human-driven Ecosystem Deviations (1989631)
Jiangtao Liu1, Chaopeng Shen1, Chonggang Xu2, Yuchen Bian3, Marc McDill4 and Nicholas Rigdeway Kraabel5, (1)Pennsylvania State University Main Campus, Department of Civil and Environmental Engineering, University Park, United States, (2)Los Alamos National Lab, Los Alamos, United States, (3)Amazon Search, Palo Alto, United States, (4)Pennsylvania State University Main Campus, Department of Ecosystem Science and Management, University Park, United States, (5)Pennsylvania State University Main Campus, Civil and Environmental Engineering Department, University Park, United States
CARE-SST: context-aware reconstruction diffusion model for sea surface temperature (1933393)
Minki Choo, Ulsan National Institute of Science and Technology, Department of Civil, Urban, Earth, and Environmental Engineering, Ulsan, South Korea, Sihun Jung, Ulsan National Institute of Science and Technology, Civil, Urban, Earth, and Environmental Engineering, Ulsan, South Korea, Jungho Im, Ulsan National Institute of Science and Technology, Ulsan, South Korea and Daehyeon Han, University of Colorado Boulder, Department of Geography, Boulder, United States
Data-driven model-agnostic evidence for Nonlocal Geomorphic Transport laws across geomorphic regimes (1976031)
Alejandro Tejedor1,2, Marcus Engsig3, Arvind Singh4, Oyku Deniz Kose5, Zijun Sun5, Efi Foufoula-Georgiou6 and Yanning Shen5, (1)University of Zaragoza, Institute for Biocomputation and Physics of Complex Systems (BIFI), Zaragoza, Spain, (2)University of California Irvine, Department of Civil and Environmental Engineering, Irvine, United States, (3)Technology Innovation Institute, Directed Energy Research Centre, Abu Dhabi, United Arab Emirates, (4)University of Central Florida, Department of Civil, Environmental & Construction, Orlando, United States, (5)University of California Irvine, Irvine, United States, (6)University of California Irvine, Civil and Environmental Engineering, Irvine, United States
Deep-learned additive data assimilation for improving weather forecast skill (1961226)
Chris Miller1, Matt Bender1, Joshua DaRosa1, Mohammad Ridhwaan Alam2, Nick Krall1, Kelsey Lieberman1, Nick Silverman3, Noah Brenowitz4, Laura Slivinski5 and Sergey Frolov5, (1)The MITRE Corporation, McLean, United States, (2)The MITRE Corporation, San Diego, United States, (3)The MITRE Corporation, Missoula, United States, (4)NVIDIA Corporation, Santa Clara, CA, United States, (5)NOAA/PSL, Boulder, United States
Development of an AI Agent for Hydrologic Modeling (1972694)
Songkun Yan1, Zhi Li2, Siyu Zhu1, Yixin Wen3, Mofan Zhang4, Mengye Chen5, Jie Cao6 and Yang Hong1, (1)University of Oklahoma, School of Civil Engineering and Environmental Science, Norman, United States, (2)Stanford University, Stanford, United States, (3)University of Florida, Department of Geography, Ft Walton Beach, FL, United States, (4)Stanford University, Stanford, CA, United States, (5)The University of Oklahoma, Center for Analysis and Prediction of Storms, Norman, United States, (6)University of Oklahoma Norman Campus, School of Computer Science, Norman, United States
Development of HiDRED (Himawari Data Rainfall Estimation using Deep learning), a real‑time satellite rainfall product for Southeast Asia, and an investigation of its applicability to rainfall–runoff analysis (1908764)
Kansei Fujimoto, Chuo University, Civil, Human and Environmental Science and Engineering Course, Graduate School of Science and Engineering, Bunkyo-ku, Tokyo, Japan and Taichi Tebakari, Chuo University, Depertment of Civil and Environmental Engineering, Bunkyo-ku,Tokyo, Japan
Efficient Kilometer-Scale Precipitation Downscaling with Conditional Wavelet Diffusion (Invited) (1950301)
Haizhao Yang, University of Maryland College Park, College Park, MD, United States, Chugang Yi, University of Maryland, College Park, College Park, United States, Weikang Qian, University of Florida, Department of Geography, Gainesville, United States, Yixin Wen, University of Florida, Department of Geography, Ft Walton Beach, FL, United States and Minghan Yu, University of Maryland College Park, Department of Mathematics, College Park, United States
Enhancing Wind Energy Forecasting via Solar Covariates and Language-Based Pattern Mining: A Chronos-Based Study in Texas (2001543)
Yuhan Fan1, Chengzhuo Huang2, Mengjie Zhang3 and Upmanu Lall3, (1)Columbia University, Earth and Environmental Engineering, New York City, United States, (2)Columbia University, Earth and Environmental engineering, New York, United States, (3)Columbia University, Earth and Environmental Engineering, New York, United States
Evaluating Machine Learning Methods for River Flow Time Series Modeling in U.S. Midwest Agricultural Watersheds (1967646)
Maral Khodadadi1, Bartosz P. Grudzinski2, Michael J. Vanni3, Elisabeth Widom1 and Vidya Samadi4, (1)Miami University, Department of Geology & Environmental Earth Science, Oxford, OH, United States, (2)Miami University, Department of Geography, Oxford, OH, United States, (3)Miami University, Department of Biology, Oxford, OH, United States, (4)Clemson University, Department of Agricultural Sciences, Clemson, SC, United States
Evaluating Machine Learning-Based Estimation of Mangrove Vegetation Indices from Integrated SAR and Optical Remote Sensing (1994263)
Mushrufa Mushreen Winey1, Nabanita Chowdhury1, Jubaer Al Samin2, Faiza Fariha Khan1, Sonia Binte Murshed1 and . Shampa1, (1)Bangladesh University of Engineering and Technology (BUET), Institute of Water and Flood Management (IWFM), Dhaka, Bangladesh, (2)Bangladesh University of Engineering and Technology (BUET), Dhaka, Bangladesh
Forecasting Landscape Changes with ConvLSTM Trained on Simulated Landscapes (1979987)
Saroj Basnet1, Aysan Hassanzadeh Bavojdan2, Alejandro Tejedor3,4, Efi Foufoula-Georgiou5 and Arvind Singh1, (1)University of Central Florida, Civil, Environmental, And Construction Engineering, Orlando, United States, (2)University of Central Florida, Department of Civil, Environmental & Construction, Orlando, FL, United States, (3)University of California Irvine, Department of Civil and Environmental Engineering, Irvine, United States, (4)University of Zaragoza, Institute for Biocomputation and Physics of Complex Systems (BIFI), Zaragoza, Spain, (5)University of California Irvine, Civil and Environmental Engineering, Irvine, United States
Forecasting Wind and Solar Energy Potential Using Time-Series Foundation Models: A Case Study in Texas (1930653)
Chengzhuo Huang1, Yuhan Fan2, Mengjie Zhang3 and Upmanu Lall3, (1)Columbia University, Earth and Environmental engineering, New York, United States, (2)Columbia University, Earth and Environmental Engineering, New York City, United States, (3)Columbia University, Earth and Environmental Engineering, New York, United States
Fusion of Multi-Source Precipitation Records via Coordinate-Based Generative Models (1855923)
Sencan Sun1, Baoxiang Pan2, Lu Li3, Xin Li4, Efi Foufoula-Georgiou5, Yanluan Lin6 and Congyi Nai2, (1)Tsinghua University, Beijing, China, (2)Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China, (3)Sun Yat-Sen University, School of Atmospheric Sciences, Guangzhou, China, (4)National Tibetan Plateau Data Center (TPDC), State Key Laboratory of Tibetan Plateau Earth System, Environment and Resources (TPESER), Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing, China, China, Beijing, China, (5)University of California Irvine, Civil and Environmental Engineering, Irvine, United States, (6)Tsinghua University, Department of Earth System Science, Beijing, China
Global CO₂ Retrievals Using Aerosol Estimates Modeled on OCO‑2 Observations (1955631)
Antong Zhang1,2, Vijay Natraj3, Jiani Yang4, Sihe Chen4 and Yuk L Yung4, (1)Brown University, Providence, RI, United States, (2)California Institute of Technology, Pasadena, United States, (3)Jet Propulsion Laboratory, California Institute of Technology, Pasadena, United States, (4)California Institute of Technology, Division of Geological and Planetary Sciences, Pasadena, United States
High-Resolution Climate Data for Arid Regions: A Generative Correction Diffusion Approach for Hydrological Applications in Arizona (1961405)
Hossein Yousefi Sohi, University of Arizona, Department of Hydrology and Atmospheric Sciences, Tucson, AZ, United States, Andrew Bennett, University of Arizona, Department of Hydrology & Atmospheric Sciences, Tucson, AZ, United States and Ali Behrangi, University of Arizona, Department of Hydrology and Atmospheric Sciences, Tucson, United States
Leveraging Passive Microwave Data (MiRS) With U-Net Architecture for Satellite Rainfall Estimation in High-Latitude Regions: A Case Study over Alaska (1902128)
Mohammad bolboli Zadeh1, Vu Dao1, Kuo-lin Hsu2, Phu Nguyen2 and Soroosh Sorooshian2, (1)University of California Irvine, Civil and Environmental Engineering, Irvine, CA, United States, (2)University of California, Irvine, Department of Civil and Environmental Engineering, Irvine, United States
Machine learning-based analysis of oil and gas methane emissions in the Gulf of Mexico using publicly available satellite data (1980356)
Pedro de Melo, Boulder, Colorado, UNITED STATES, Daniel J. Varon, Harvard University, School of Engineering and Applied Sciences, Cambridge, United States, Nikolay Balashov, USRA/NASA Goddard Space Flight Center, Greenbelt, United States, Ryan M Stauffer, NASA Goddard Space Flight Center, Greenbelt, United States, Daven K Henze, University of Colorado Boulder, Boulder, United States and Lori Bruhwiler, NOAA, Global Monitoring Laboratory, Boulder, United States
Mapping and Monitoring Reservoir Dynamics with a Fine-Tuned Foundation Deep Learning Model and Time-Series SAR Data (1850164)
Yitao Pu1, Hongxing Liu1, Shujie Wang2, Haibin Su3, Lei Wang4 and Matthew C LaFevor5, (1)The University of Alabama, Department of Geography and the Environment, Tuscaloosa, AL, United States, (2)Pennsylvania State University, Geography, Earth and Environmental Systems Institute, ICDS, University Park, PA, United States, (3)Texas A&M University - Kingsvi, Department of Physics and Geosciences, Kingsville, United States, (4)Louisiana State University, Department of Geography and Anthropology, Baton Rouge, LA, United States, (5)University of Alabama, Department of Geography and the Environment, Tuscaloosa, AL, United States
Modeling Flood Susceptibility Through Explainable Artificial Intelligence Techniques: Case study in Beni Mellal Region of Morocco (1993346)
Hamza Legsabi1, Soufiane Taia2, Sidi Mohamed Boussabou1, Nora Nejaoui3, Bouabid El Mansouri1 and Lamia Erraioui1, (1)Natural Resources and Durable Development Laboratory, Science and Technical Faculty, Ibn Tofail University, Kenitra, Morocco, (2)Laboratory of Applied Geology and Geo-Environment, Ibn Zohr University, Geology, Agadir, Morocco, (3)Pollution Control Department, Water Research and Planning Directorate, Rabat, Morocco
NeRF for 3D Scene Construction from Satellite Imagery of Polar Regions (1902728)
Ali Ozdener, University of Georgia, Athens, GA, United States, Ellemieke Van Kints, University of Georgia, Athens, United States, Nancy O'Hare, University of Georgia, Department of Geography, Athens, United States and Deepak R Mishra, University of Georgia, Center for Geospatial Research, Department of Geography, Athens, United States
Neural Radiance Field (NeRF) for 3D Reconstruction from Drone Imagery (1869254)
Evan Hammam, University of Georgia, Athens, GA, United States, Nancy O'Hare, University of Georgia, Department of Geography, Athens, United States and Deepak R Mishra, University of Georgia, Center for Geospatial Research, Department of Geography, Athens, United States
Performance of Satellite-Derived Precipitation Products over the Santa Ana River Basin for Reservoir Operations at Prado Dam (1901253)
Kritan Subedi1, Phu Nguyen2 and Soroosh Sorooshian2, (1)University of California Irvine, Irvine, CA, United States, (2)University of California, Irvine, Department of Civil and Environmental Engineering, Irvine, United States
Physics-informed bias correction using a convolution-based multivariate Gaussian process (1913158)
Taereem Kim1, Chao Wang2, Gabriele Villarini3, James Done4, David Johnson5 and Andreas F Prein4,6, (1)Princeton University, Civil and Environmental Engineering & High Meadows Environmental Institute, Princeton, NJ, United States, (2)University of Iowa, Department of Industrial and Systems Engineering, Iowa City, United States, (3)Princeton University, Civil and Environmental Engineering & High Meadows Environmental Institute, Princeton, United States, (4)NSF National Center for Atmospheric Research, Boulder, United States, (5)Purdue University, Industrial Engineering & Political Science, West Lafayette, IN, United States, (6)ETH Zurich, Institute of Atmospheric and Climate Science, Zurich, Switzerland
Predicting present and climate change impacts on groundwater nitrate pollution and risk in India at high resolution. (2002654)
Pabitra Senapati, Indian Institute of Technology Kharagpur, Geology & Geophysics, Kharagpur, India, Soumyajit Sarkar, Indian Institute of Technology Kharagpur, Kharagpur, India and Abhijit Mukherjee, Indian Institute of Technology Kharagpur, Department of Geology and Geophysics; School of Environmental Science and Engineering, Kharagpur, India
Prediction of Local Scour Evolution Beneath a SubmergedCylinder by U-Shaped Fourier Neural Operators (1988825)
Haq Nazari, Louisiana State University, Baton Rouge, LA, United States, Celalettin E Ozdemir, Louisiana State University, Civil & Environmental Engineering, Baton Rouge, LA, United States and Mayank Tyagi, Louisiana State University, Baton Rouge, United States
Sedimentological Characteristics Using Machine Learning Ensembles for Rock Avalanche Susceptibility Modelling of Skardu Basin, Northern Pakistan. (1867120)
Surih Sibaghatullah Jagirani, University of Sindh Jamhsoro, Sindh, Centre For Pure And Applied, Jamshoro, Sindh, Pakistan, Irshad Ali Zardari, IMHE Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, State Key Laboratory of Mountain Hazards and Engineering Resilience, Chengdu, Sichuan, China, Muhammad Hassan Agheem, University of Sindh Jamshoro, Centre for Pure and Applied Geology, Jamshoro, Sindh, Pakistan, Weiming Liu, CAS Key Laboratory of Mountain Hazards and Surface Process, Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Chengdu, China and Liu Weiming, Nafees Ali, Hamza Daud , Hamid Faiz
Self-Organizing Maps to define Wet Season Onset over the Amazon Rainforest and Analyze Long-term Trends in Atmospheric Radiation and Precipitation (1971999)
Siddhant Gupta, Argonne National Laboratory, Argonne, IL, United States, Eunice Ledres, State University of New York New Paltz, New Paltz, United States and Dié WANG, Brookhaven National Laboratory, Upton, NY, United States
Spatial Assessment of Two Satellite Rainfall Products Using Quality Controlled Ground-Based Interpolation in Saudi Arabia: Revealing the Need for Regional Bias Correction (1902623)
Naif Aljehani1, Phu Nguyen2 and Soroosh Sorooshian2, (1)University of California Irvine, Irvine, CA, United States, (2)University of California, Irvine, Department of Civil and Environmental Engineering, Irvine, United States
Synthetic Storm Event Generation Using Stable Diffusion to Improve Power Outage Prediction (1894489)
Morteza Azizi1, Xinxuan Zhang1, Dongjin Song2, Diego Cerrai3 and Emmanouil N Anagnostou4, (1)University of Connecticut, Civil and Environmental Engineering, Storrs, United States, (2)University of Connecticut, Computer Science and Engineering, Storrs, CT, United States, (3)University of Connecticut, School of Civil and Environmental Engineering, Storrs, United States, (4)Professor, Civil and Environmental Engineering, Storrs, United States
Temporal Transfer Learning for Out-of-distribution Generalization in Climate Downscaling (1872916)
Shuochen Wang, Northeastern University, Civil and Environmental Engineering, Boston, MA, United States, Nishant Yadav, Microsoft Corporation, Redmond, United States and Auroop Ganguly, Northeastern University, Civil and Environmental Engineering, Boston, United States
Transfer Learning with Recurrent Neural Networks of Fuel Moisture Content: Extending from 10h fuels to 1h and 100h (1954155)
Jonathon Hirschi, University of Colorado Denver, Denver, CO, United States, Jan Mandel, University of Colorado Denver, Mathematical and Statistical Sciences, Denver, CO, United States and Angel Farguell, University of Colorado Denver, Mathematical and Statistical Sciences, Denver, United States
Understanding and Predicting Flash Flood Impact Locations Using Machine Learning and Radar Rainfall Data (1920506)
Paniz Peiravani1, Yingli Tian2, Te Pei3, Seonho Kim4, Jae-Ung Yu4 and Naresh Devineni4, (1)Department of Earth System Science & Environmental Engineering, The City College of the City University of New York, and CUNY-CREST, Department of Earth System Science & Environmental Engineering, New York, United States, (2)Department of Electrical Engineering, The City College of the City University of New York, and CUNY-CREST, Department of Electrical Engineering, New York, United States, (3)Department of Civil Engineering, Stony Brook University, Department of Civil Engineering, New York, United States, (4)Department of Civil Engineering, The City College of the City University of New York, and CUNY-CREST, Department of Civil Engineering, New York, United States
