SOIL MOISTURE ESTIMATION OF EUCALYPTUS FORESTS IN PORTUGAL WITH L-BAND SAR USING POLARIMETRIC DECOMPOSITIONS AND MACHINE LEARNING - Dissertation

Aug 5, 2026

Upload by Himbo

Soil moisture is a critical ecological parameter driving complex land–atmosphere interactions. Remote sensing with microwaves offers continuous monitoring potential, yet Mediterranean Portugal’s low and sporadic precipitation and non-native eucalyptus plantations complicate water availability assessment. This study uses fully polarimetric L-band SAR and machine learning to predict soil moisture in eucalyptus forests and compare it with other forest types. Four polarimetric decompositions and additional radar indices provide features, followed by feature selection and cross-validated retrieval. Random Forest achieves the best retrieval performance (RMSE 0.021, MAE 0.017, MBE 0.001). A t-test indicates no significant difference in predicted soil moisture between eucalyptus and other forests in the study area.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages75

Reading Time3 h 9 min

AccessFree to download

Abstract

Keywords

Acronyms

SAR and polarization terms

Indices, models, and evaluation metrics

SOIL MOISTURE ESTIMATION OF EUCALYPTUS FORESTS IN PORTUGAL WITH L-BAND SAR USING POLARIMETRIC DECOMPOSITIONSAND MACHINE LEARNING Dissertation supervised by: Joel Dinis Baptista Ferreira da Silva PhD, Universidade Nova de Lisboa, IMS Hugo Alexandre Gomes da Costa PhD, Universidade Nova de Lisboa, IMS Carlos Granell Canut PhD, Universitat Jaume I, INIT February 2023 DECLARATION OF ORIGINALITY I declare that the work described in this document is my own and not from someone else. All the assistance I have received from other people is duly acknowledged and all the sources (published or not published) are referenced. This work has not been previously evaluated or submitted to NOVA Information Management School or elsewhere. Daniel Erickson 24.02.2023 Lisbon The signed original has been archived by NOVA IMS. ACKNOWLEDGMENTS To my parents, who have supported and guided me to get the best education possible, to my partner, who is my daily emotional support, to my advisors and professors, who helped me to reach this great achievement, to the other students of the GeoTech Master program, who have been true friends, to my friends and other family members, who have encouraged me, to Dr. Marco Painho, who helped innumerable times throughout the program, and to 8 year-old me, who only dreamt of reaching this. SOIL MOISTURE ESTIMATION OF EUCALYPTUS FORESTS IN PORTUGAL WITH L-BAND SAR USING POLARIMETRIC DECOMPOSITIONSAND MACHINE LEARNING ABSTRACT Soil moisture is a critical ecological parameter because it is a primary input for all processes that involve the complex interaction between land surface and the atmosphere. Remote sensing, especially using microwaves, has shown great promise in measuring soil moisturewith several operating satellites focused on its continuous estimation and monitoring on a global scale. Portugal is predominantly characterized by Mediterranean and semi-arid climates that feature low and sporadic precipitation. Over 10% of Portugal’s land area has been planted with Eucalyptus globulus-a non-native, fast-growing tree primarily planted for industrial use. Some studies have demonstrated that eucalyptus plantations adversely affect water availability, but overall results have been inconclusive as there are numerous other confounding variables. The goals of this study were to determine, using fully polarimetric L-band SAR and machine learning, if soil moisture could be accurately predicted in eucalyptus forests, and if there is a significant difference in soil moisture inside eucalyptus forests relative to other forests. Vegetated surfaces complicate the estimation of soil moisture because their structure and water content contribute significantly to backscatter of the radar signal. Thus, four polarimetric decompositions were compared to separate vegetative versus surface backscatter. The inputs from those decompositions, as well as several additional radar indices and polarizations from the microwave images, were used as feature inputs into two different machine learning models. After a feature selection process, the soil moisture estimations were retrieved and compared using cross-validation. The best overall soil moisture retrieval for Eucalyptus forests came from Random Forest with a RMSE of 0.021, a MAE of 0 .017, and a MBE of 0 .001. Through a statistical t-test, predicted soil moisture values in eucalyptus forests did not differ significantly as compared to other forest types in the study area. KEYWORDS Soil Moisture Polarimetry Eucalyptus L-Band SAR Feature Selection Machine Learning ACRONYMS ALOS –Advanced Land Observing Satellite (Japan) ASCAT –Advanced Scatterometer (European Space Agency) ASI –Agenzia Spaziale Italiana (Italian Space Agency) BIOMASS – Biomass Monitoring Mission for Carbon Assessment (European Space Agency) [C] – Covariance Matrix CONAE – Comisión Nacional de Actividades Espaciales (Argentinian Space Agency) COSMO-SkyMed – Constellation

FAQs

Why is soil moisture estimation important in this study?

Soil moisture is a key ecological parameter that influences land–atmosphere processes. Accurate estimation supports understanding water-related dynamics in forest ecosystems.

What data and methods are used to predict soil moisture?

The study uses fully polarimetric L-band SAR imagery and machine learning. Four polarimetric decompositions are compared to separate vegetative versus surface backscatter and provide feature inputs.

How accurate is the best soil moisture retrieval model?

The best overall performance for eucalyptus forests comes from Random Forest, with RMSE 0.021, MAE 0.017, and MBE 0.001 after feature selection and cross-validation.

Aug 5, 2026

Upload by Himbo