Streamflow and baseflow generation
Understanding how recharge, infiltration, and runoff pathways shape low-flow realism in dry regions.
My projects connect physical hydrology, climate analytics, and scientific computing. This page groups them into a clearer structure so visitors can quickly understand the themes behind the individual case studies.
Understanding how recharge, infiltration, and runoff pathways shape low-flow realism in dry regions.
Studying how land models represent memory, retention, and land-atmosphere interactions over time.
Packaging methods into software that supports repeatable analysis and broader research reuse.
Examining drought, precipitation intensity, and resilience across hydrologic systems.
Designing interpretable ML architectures for routing, rainfall-runoff prediction, and hybrid physical plus data-driven modeling.
Building dashboards, statistical analyses, and data-management tools that turn environmental data into decisions.
A deeper look at how soil hydraulic representation affects persistence, climate feedbacks, and model performance.
Open SMM case study →An investigation of baseflow generation mechanisms and their consequences for streamflow prediction in the arid Southwest.
Open BFI case study →A new graph-based routing framework that uses directed river-network topology and Noah-MP runoff to improve streamflow routing skill.
Open RiverGraphNet page →A process-aware mass-conserving neural framework for rainfall-runoff modeling with interpretable storage-flux dynamics.
Open SoilSolver page →A sequence-modeling project comparing LSTM and Transformer approaches for streamflow prediction, with a live dashboard for result exploration.
Open Streamflow Prediction page →Probabilistic estimation projects using MCMC and multi-view geometry show my comfort with inference, uncertainty, and model structure.
Prediction and classification work on crash severity, climate trends, and neural-network benchmarks show a broader ML toolkit beyond hydrology.
Interactive dashboards and analytical summaries translate raw data into interpretable patterns for public safety and planning contexts.
SQL and smart water distribution work demonstrate database design, query logic, and system thinking for operational water problems.