Reiman Alfonso has been a doctoral fellow at the National Scientific and Technical Research Council (CONICET) since 2023 and is a doctoral student in engineering at the Faculty of Engineering and Water Sciences (FICH) of the National University of the Littoral (UNL) in Santa Fe, Argentina. He holds a Master’s degree in Advanced Computer Science (2021) from the Faculty of Interactive Technologies at the University of Computer Science (UCI), Havana, Cuba, and a degree in Computer Science Engineering (2017) from the same institution. He is currently working on the application of deep learning and remote sensing techniques for monitoring environmental variables, with an emphasis on estimating soil moisture and studying processes associated with fire in wetlands, carrying out his activities at the Center for River and Hydro-Environmental Studies of the Littoral (CEFHAL).

The talk will address high-resolution soil moisture estimation, a key variable for agricultural and hydrological applications, whose satellite monitoring remains a challenge. An approach based on a U-Net convolutional neural network will be presented, which allows soil moisture to be estimated at 30 m resolution from Landsat 8 images in the Pampas region of Argentina. The model, trained with data from the Argentine SAOCOM mission, performs well when validated with in situ measurements and allows the reconstruction of a continuous high-resolution time series for the period 2013–2025, with potential application in the analysis of droughts and water stress at the regional scale.

The seminar will take place on Friday, January 9 ^th at 12:00 PM Chile time (3:00 PM UTC) in the Maryam Mirzakhani Room at the Center for Mathematical Modeling, University of Chile (Beauchef 851, North Building, 6th floor). It will also be live-streamed at this link.