H138
Utility of Artificial Intelligence/Machine Learning Approaches in Soil Hydrological Processes and Agriculture

Session ID#: 74487

Session Description:
While improving our understanding of multi-scale process drivers in the vadose zone is crucial in the long term, it is also necessary to consider short term solutions to current issues in efficient agriculture and water management. Rapid strides are being taken towards adoption of machine learning and artificial intelligence methods to improve our predictive capabilities for water and nutrient dynamics in the unsaturated zone. With agriculture being the major consumer and significant source of contamination of freshwater resources,  the deluge of data from newer remote sensing platforms provides an unprecedented opportunity for the scientific community to understand the underlying soil hydrological processes and its impact in agriculture. This session invites contributions describing the applications using artificial intelligence, machine learning and/or other data science-based techniques towards understanding and predicting the flow of water and nutrients in the vadose zone across space and time scales, and its impacts and management in agriculture.
Index Terms:

1805 Computational hydrology [HYDROLOGY]
1849 Numerical approximations and analysis [HYDROLOGY]
1875 Vadose zone [HYDROLOGY]
1942 Machine learning [INFORMATICS]

Primary Convener:  Raghavendra B Jana, Skolkovo Institute of Science and Technology, CDISE, Skolkovo, Russia
Conveners:  Amor V M Ines, Michigan State University, East Lansing, MI, United States and Yuying Xie, Michigan State University, Computational mathematics, Science and Engineering, East Lansing, United States
Primary Liaison:  Raghavendra B Jana, Skolkovo Institute of Science and Technology, CDISE, Skolkovo, Russia
 
Soil Moisture Characteristics Curve Measurements in a Sandy Loam as Affected by Tillage (487108)
Jay Jabro, USDA-ARS, Sidney, MT, United States and Bart W. Stevens, USDA, Sidney, United States
 
Near-real-time forecast of satellite-based soil moisture using time series deep learning models (Invited) (492168)
Chaopeng Shen and Kuai Fang, Pennsylvania State University Main Campus, Civil and Environmental Engineering, University Park, PA, United States
 
Aiding land and water management in agricultural areas: Machine learning methods predict and map shallow aquifer redox conditions and indicate denitrification potential in an important extensive unconsolidated drinking water aquifer system in the northern agricultural areas of the U.S. (496842)
Melinda L Erickson, U.S. Geological Survey, Upper Midwest Water Science Center, Minneapolis, MN, United States, Sarah M. Elliott, USGS Upper Midwest Water Science Center, Mounds View, MN, United States, Paul E. Stackelberg, USGS New York Water Science Center Troy, Troy, NJ, United States, Craig Brown, USGS New England Water Science Center, Northborough, United States and Katherine Marie Ransom, U.S. Geological Survey, Sacramento, CA, United States
 
Restoration of Landsat-7 images using Deep Image Prior (505102)
Raghavendra B Jana, Anna Petrovskaia and Ivan V Osedledets, Skolkovo Institute of Science and Technology, CDISE, Skolkovo, Russia
 
Deep Learning on Time Series of Remote Sensing Data to Predict the Occurrence of Soil Water Repellency in Dryland Pastures (514059)
Mohamed Bayad1, Henry Chau1, Stephen Trolove2, Leo Condron3, Jim Moir1 and Yi Li4, (1)Lincoln University, Christchurch, New Zealand, (2)New Zealand Institute of Plant and Food Research, Auckland, New Zealand, (3)Lincoln Univeresity, Christchurch, New Zealand, (4)Northwest Agriculture and Forestry University, Yangling, China
 
Random forest to predict and forecast subsurface soil moisture in agricultural fields (550883)
Coleen Carranza, Wageningen University and Research Center, Wageningen, Netherlands and Martine J van der Ploeg, Wageningen University, Wageningen, Netherlands
 
A Machine Learning Approach to Retrieve Soil Textural Information Using Soil Moisture Dynamics (587716)
Rajsekhar Kandala1, Sat Kumar Tomer2, Dharumarajan S3, Raghavendra B Jana4 and Sekhar Muddu1, (1)Indian Institute of Science, Department of Civil Engineering, Bangalore, India, (2)Satyukt Analytics Pvt Ltd, Bangalore, India, (3)ICAR-National Bureau of Soil Survey and Land Use Planning, Bangalore, India, (4)Skolkovo Institute of Science and Technology, CDISE, Skolkovo, Russia
 
3D representation of soil structure using Generative Adversarial Networks (587735)
Anna Petrovskaia and Raghavendra B Jana, Skolkovo Institute of Science and Technology, CDISE, Skolkovo, Russia
 
Remote Sensing Precipitation for Agricultural Applications using Machine-Learning Methods (Invited) (604701)
Kuo-lin Hsu, University of California Irvine, Irvine, CA, United States
 
Pattern extraction of soil heterogeneity and soil-crop interaction using unsupervised Bayesian learning: an application to satellite-derived NDVI time series and electromagnetic induction measurements (621987)
Hui Wang, University of Dayton, Dayton, OH, United States, Florian Wellmann, RWTH Aachen University, Aachen, Germany and Jan Van Der Kruk, Forschungszentrum Jülich, Agrosphere (IBG-3), Institute of Bio- and Geosciences, Jülich, Germany
See more of: Hydrology
<< Previous Session Proposal | Next Session Proposal >>