H002
Advances and Applications of Data Integration, Inverse Methods, and Machine Learning in Hydrogeophysics
Session ID#: 71766
Session Description:
The combined use of geophysical and hydrogeological measurements in groundwater investigations have been shown to improve the predictive accuracy of groundwater models, aid in uncertainty assessment of hydrogeologic parameters, and enhance the monitoring and understanding of hydrogeologic processes. This session solicits research that incorporates geophysical and/or hydrogeological measurements to improve the characterization and monitoring of hydrogeologic parameters and processes. Topics of interest include, but are not limited to: innovative coupled or joint inversion strategies; machine learning strategies to incorporate hydrogeologic process information into geophysical inverse procedures; data errors and uncertainty quantification methods; proxy-modeling strategies; case studies of incorporation of geophysical methods in groundwater investigations.
Index Terms:
0555 Neural networks, fuzzy logic, machine learning [COMPUTATIONAL GEOPHYSICS]
1829 Groundwater hydrology [HYDROLOGY]
1835 Hydrogeophysics [HYDROLOGY]
1873 Uncertainty assessment [HYDROLOGY]
Primary Convener: Erasmus K. Oware, University at Buffalo, Buffalo, NY, United States
Conveners: Emily Voytek, University of Lausanne, Lausanne, Switzerland, Thomas Hermans, University of Liege, Urban and Environmental Engineering, Liege, Belgium and Niels Grobbe, University of Hawaii at Manoa, Hawaii Institute of Geophysics and Planetology, Honolulu, HI, United States
Primary Liaison: Erasmus K. Oware, University at Buffalo, Buffalo, NY, United States
Deep learning and inversion (Invited) (487616)
Niklas Linde1, Eric Laloy2, Jürg Hunziker3 and James Irving1, (1)University of Lausanne, Lausanne, Switzerland, (2)Belgian Nuclear Research Centre (SCK-CEN), Mol, Belgium, (3)University of Lausanne, Department of Earth Sciences, Lausanne, Switzerland
Improving Bayesian Evidential Learning 1D imaging (BEL1D) accuracy through iterative prior resampling (496147)
Hadrien Michel, F.R.S.-FNRS (Fonds de la Recherche Scientifique), Brussels, Belgium; University of Liège, Urban and Environmental Engineering, Liège, Belgium, Frederic Nguyen, University of Liege, Urban and Environmental Engineering, Liege, Belgium and Thomas Hermans, Ghent University, Department of Geology, Ghent, Belgium
Determination of the correlation structure of an alluvial aquifer from multi-frequency 3D GPR reflection measurements (499596)
James Irving1, Zhiwei Xu1,2, Kyle M Lindsay3, John Bradford4, Peimin Zhu2 and Klaus Holliger1,5, (1)University of Lausanne, Lausanne, Switzerland, (2)China University of Geosciences, Wuhan, China, (3)Boise State University, Boise, ID, United States, (4)Colorado School of Mines, Golden, CO, United States, (5)Zhejiang University, Hangzhou, China
Time-lapse seismic and electrical monitoring of the vadose zone during a controlled infiltration experiment at the Ploemeur Hydrological Observatory (Brittany, France) (501127)
Lara Antonia Blazevic1, Ludovic Bodet1, Niklas Linde2, Laurent Longuevergne3, Sylvain Pasquet4, Thomas Hermans5 and Damien Jougnot1, (1)Sorbonne University, UMR 7619 METIS, Paris, France, (2)University of Lausanne, Lausanne, Switzerland, (3)Geosciences Rennes, UMR 6118 CNRS - University of Rennes, Rennes, France, (4)Institut de Physique du Globe de Paris, Paris, France, (5)Ghent University, Department of Geology, Ghent, Belgium
Direct forecasting of local hydraulic conductivity using combined geophysical and hydrological data: Application to well catchment predictions in Danish aquifer system (502994)
Lijing Wang, Stanford University, Stanford, CA, United States, Troels N Vilhelmsen, Aarhus University, Department of Geoscience, Aarhus, Denmark and Jef Caers, Stanford University, Department of Geological Sciences, Stanford, CA, United States
Time-lapse Full-waveform Inversion of Dynamic Geophysical Processes with Examples from Multi-phase Flow (572046)
Dongzhuo Li, Stanford University, Department of Geophysics, Stanford, CA, United States, Kailai Xu, Stanford University, Institute for Computational and Mathematical Engineering, Stanford, United States, Jerry M Harris, Stanford University, Stanford, CA, United States and Eric F Darve, Stanford University, Mechanical Engineering, Stanford, CA, United States
An Approximate Fast-Mapping Approach to the Surface NMR Forward Problem (573231)
Denys Grombacher, Aarhus University, Department of Geoscience, Aarhus C, Denmark, Andy Kass, Aarhus University, Department of Geoscience, Aarhus, Denmark, Esben Auken, University of Aarhus, Aarhus, Denmark and Jakob Juul Larsen, Aarhus University, Aarhus C, Denmark
Efficient Deterministic Inversion of Geophysical Data Constrained by Deep Generative Models to Estimate Subsurface Parameters with Defined Spatial Patterns (575458)
Jorge Lopez-Alvis1,2, Eric Laloy3, Frederic Nguyen2 and Thomas Hermans1, (1)Ghent University, Department of Geology, Ghent, Belgium, (2)University of Liege, Urban and Environmental Engineering, Liege, Belgium, (3)Belgian Nuclear Research Centre (SCK-CEN), Mol, Belgium
Discrete Fracture Network reconstruction using a tomographic approach (577621)
Peter Bayer1, Lisa M. Ringel2, Mohammadreza Jalali3, Andreas Englert1 and Mark Somogyvari4, (1)University of Halle-Wittenberg, Halle, Germany, (2)Ingolstadt University of Applied Sciences, Ingolstadt, Germany, (3)RWTH Aachen, Aachen, Germany, (4)Technical University of Berlin, Berlin, Germany
Helicopter electromagnetic informed learning to estimate uncertainty in subsurface models (619814)
Michael J Friedel, Pacific Northwest National Laboratory, Richland, WA, United States, Burke J Minsley, USGS Geology, Geophysics, and Geochemistry Science Center, Denver, CO, United States and Lucas P Moreira, Catholic University of Brasilia, Civil Engineering Department, Brasilia, Brazil
