DONATE

IAHR Document Library


« Back to Library Homepage « Book of Abstracts of the 16th International Conference on Hy...

Estimating Root-Zone Soil Moisture with Machine Learning and Remote Sensing Data

Author(s): Judith Cid-Gimenez; Maria Jose Escorihuela; Anais Barella-Ortiz; Pere Quintana-Segui

Linked Author(s):

Keywords: Agricultural drought; Machine learning; Remote sensing; Root-zone soil moisture

Abstract: This study estimates root-zone soil moisture (RZSM) in vineyards in the Terra Alta county (Catalonia, Spain) using a multilayer perceptron (MLP). The model predicts daily RZSM from in-situ surface soil moisture, meteorological variables, and soil properties. When trained and tested on independent years at the same stations, it achieves high skill (median KGE = 0.9). In leave-station-out experiments, performance remains good at some unseen stations (KGE ≈ 0.85) but decreases at others due to biases, highlighting the need to improve spatial transferability.

DOI:

Year: 2026

Copyright © 2026 International Association for Hydro-Environment Engineering and Research. All rights reserved. | Terms and Conditions