Enhancing Hydrologic Model Reliability Through Earth Observation Integration: Tool Development, Multi-Variable Evaluation, and Scale-Aware Data Assimilation

Accurately representing the complexity of hydrologic processes remains a central challenge in water resources modeling, particularly due to limitations in traditional evaluation methods that rely heavily on streamflow alone. This dissertation addresses this gap by exploring the integration of Earth Observations (EO)—specifically evapotranspiration (ET), soil moisture (SM), and leaf area index (LAI)—into hydrologic modeling to enhance model reliability and internal process consistency. First, a novel tool, HydroMET-EO, was developed as the first desktop-based, semi-automatic software designed to bridge interoperability barriers between EO datasets and hydrologic model outputs. Demonstrated through case studies, HydroMET-EO enabled detailed sub-watershed model evaluations, revealing discrepancies not captured by streamflow-only assessment. Second, multi-variable model evaluation using EO-based ET and SM across 15 basins confirmed that strong streamflow performance does not necessarily equate to accurate representation of hydrologic processes. The results highlighted that EO integration improves diagnostic insight, particularly in ungauged areas. Third, the effectiveness of EO data assimilation was examined across three diverse basins using PET and LAI, revealing that spatial and temporal scales of assimilation must be tailored to watershed characteristics for optimal results. Improvements in streamflow, AET, and SM were most pronounced at yearly temporal scales and finer spatial resolutions in ecohydrologically sensitive basins. Collectively, the research contributes a practical evaluation tool, empirical evidence supporting EO-informed validation, and strategic guidance on data assimilation, advancing the field toward more transparent, robust, and process-consistent hydrologic modeling. These findings promote the integration of EO into mainstream modeling workflows, supporting both scientific progress and water resource decision-making.