Imaging Hyporheic Exchange by Integrating Deep Learning and Physics‐Informed Inversion of Time‐Lapse Self‐Potential Data

محفوظ في:
التفاصيل البيبلوغرافية
الحاوية / القاعدة:Geophysical Research Letters vol. 52, no. 21 (Nov 16, 2025)
المؤلف الرئيسي: Yin, Huichao
مؤلفون آخرون: Ikard, Scott J., Rucker, Dale F., Brooks, Scott C., Dai, Zhenxue, Carroll, Kenneth C.
منشور في:
John Wiley & Sons, Inc.
الموضوعات:
الوصول للمادة أونلاين:Citation/Abstract
Full Text
Full Text - PDF
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!

MARC

LEADER 00000nab a2200000uu 4500
001 3268733488
003 UK-CbPIL
022 |a 0094-8276 
022 |a 1944-8007 
024 7 |a 10.1029/2025GL118772  |2 doi 
035 |a 3268733488 
045 0 |b d20251116 
084 |a 107510  |2 nlm 
100 1 |a Yin, Huichao  |u Plant & Environmental Sciences Department, New Mexico State University, Las Cruces, NM, USA 
245 1 |a Imaging Hyporheic Exchange by Integrating Deep Learning and Physics‐Informed Inversion of Time‐Lapse Self‐Potential Data 
260 |b John Wiley & Sons, Inc.  |c Nov 16, 2025 
513 |a Journal Article 
520 3 |a Self‐potential (SP) monitoring is increasingly used for subsurface flow characterization due to its sensitivity to hydrogeological and geochemical processes. However, SP inversion remains challenging due to its ill‐posed nature, sparse data coverage, and strong transient noise. This study proposes a hybrid framework to image hyporheic exchange using a time‐lapse SP data set monitored from a streamflow site in Oak Ridge, Tennessee. Dipole moment tomography grids generated from the physics‐informed numerical inversion is first used to train a Vision Transformer (ViT) model that maps surface SP sequences to 2D source distributions. While the numerical method is more responsive to transient signals, the ViT model better captures persistent spatial structures. Their complementary outputs are jointly analyzed in the spatiotemporal domain to isolate dynamic hyporheic exchange zones and distinguish transient from steady state subsurface flow features. This approach integrates physical inversion and deep learning to enhance interpretability, generalization, and temporal awareness in SP analysis. 
653 |a Geology 
653 |a Tomography 
653 |a Surface water 
653 |a Datasets 
653 |a Deep learning 
653 |a Exchanging 
653 |a Physics 
653 |a Optimization 
653 |a Data assimilation 
653 |a Stream flow 
653 |a Numerical analysis 
653 |a Hydrology 
653 |a Image processing 
653 |a Numerical methods 
653 |a Floodplains 
653 |a Hydrogeology 
653 |a Groundwater 
653 |a Equilibrium flow 
653 |a Subsurface flow 
653 |a Methods 
653 |a Dipole moments 
653 |a Algorithms 
653 |a Stream discharge 
653 |a Mathematical models 
653 |a Environmental 
700 1 |a Ikard, Scott J.  |u U.S. Geological Survey, Oklahoma‐Texas Water Science Center, Austin, TX, USA 
700 1 |a Rucker, Dale F.  |u hydroGEOPHYSICS, Inc., Tucson, AZ, USA 
700 1 |a Brooks, Scott C.  |u Oak Ridge National Laboratory, Oak Ridge, TN, USA 
700 1 |a Dai, Zhenxue  |u Institute of Intelligent Simulation and Early Warning for Subsurface Environment, Jilin University, Changchun, China 
700 1 |a Carroll, Kenneth C.  |u Plant & Environmental Sciences Department, New Mexico State University, Las Cruces, NM, USA 
773 0 |t Geophysical Research Letters  |g vol. 52, no. 21 (Nov 16, 2025) 
786 0 |d ProQuest  |t Science Database 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/3268733488/abstract/embedded/L8HZQI7Z43R0LA5T?source=fedsrch 
856 4 0 |3 Full Text  |u https://www.proquest.com/docview/3268733488/fulltext/embedded/L8HZQI7Z43R0LA5T?source=fedsrch 
856 4 0 |3 Full Text - PDF  |u https://www.proquest.com/docview/3268733488/fulltextPDF/embedded/L8HZQI7Z43R0LA5T?source=fedsrch