Banca de DEFESA: ALEJANDRA BUSTILLOS VEGA

Uma banca de DEFESA de DOUTORADO foi cadastrada pelo programa.
DISCENTE: ALEJANDRA BUSTILLOS VEGA
DATA: 06/03/2026
HORA: 14:00
LOCAL: Google Meet
TÍTULO:

Data-Driven Hydrology: Rainfall Modeling in Forest Ecosystems Using Machine Learning

PALAVRAS-CHAVES:

Data-driven hydrology; rainfall reconstruction; machine learning; forest interception; ecohydrology; tropical forests

PÁGINAS: 120
GRANDE ÁREA: Ciências Agrárias
ÁREA: Engenharia Agrícola
SUBÁREA: Engenharia de Água e Solo
ESPECIALIDADE: Conservação de Solo e Água
RESUMO:

This study develops an Integrated Methodological Framework based on Machine Learning (ML) for the reconstruction, modeling, and transformation of rainfall in seasonal forest ecosystems, jointly addressing data availability, hydrological variability, and the fundamental ecohydrological processes controlling water input to the soil. First, a robust methodology is proposed for reconstructing continuous hourly and daily rainfall series in regions with incomplete monitoring networks. The approach combines automated quality control through unsupervised anomaly detection, bias correction of satellite and reanalysis products, and optimal model identification based on multi-objective criteria that balance statistical accuracy and hydrological consistency. Results show that ML models trained on quality-controlled data significantly outperform traditional statistical methods while preserving convective variability, which is essential for hydrological applications in tropical ecosystems. The study advances rainfall modeling by explicitly incorporating forest canopy interception processes. Using Machine Learning ensembles and a set of hydrologically interpretable predictors (event magnitude, antecedent moisture, seasonality, and spatial heterogeneity), the results demonstrate strong spatiotemporal variability in interception, which cannot be adequately represented by constant coefficients. The models capture nonlinear relationships between rainfall, canopy structure, and hydrological memory, enabling more realistic estimates of the fraction of precipitation reaching the soil. Overall, the study shows that rainfall reconstruction and its transformation into effective precipitation are inseparable processes in the ecohydrology of seasonal forests. The use of Machine Learning within physically informed and rigorously validated frameworks improves the representation of key hydrological processes and provides reproducible tools for hydrological cycle analysis, ecohydrological modeling, and water resources management in tropical forest ecosystems.

MEMBROS DA BANCA:
Interno - GILBERTO COELHO (Membro)
Externo à Instituição - JOSÉ ALVES JUNQUEIRA JÚNIOR - IFMG (Suplente)
Externo à Instituição - SAMUEL BESKOW - UFPel (Membro)
Interno - MARCELO RIBEIRO VIOLA (Suplente)
Externo à Instituição - GERNOT HEISENBERG - TH Köln (Membro)
Presidente - CARLOS ROGERIO DE MELLO (Membro)
Interno - ANDRÉ FERREIRA RODRIGUES - UFMG (Membro)