HydroBlox: AI-Assisted Visual Programming Framework for Enhanced Scientific Reproducibility in Hydrology
Day 2 Session 2 (1:30PM)
Presenters:
Carlos Erazo (Tulane University)
Ibrahim Demir (Tulane University)
This workshop introduces students and researchers to HydroBlox, a web-based, no-code, plug-and-play hydrological research workflow platform. Participants will learn the complete lifecycle of a data project—from initial data acquisition and browser-based coding to advanced analysis and insight generation.
The session culminates in the creation of a standalone, exportable application, providing attendees with a comprehensive toolkit for modern water data workflows and improved scientific reproducibility in hydrology.
Learning Outcomes:
- Web-based tools for hydrological workflow orchestration
- Executing Python, JavaScript, and WebAssembly directly in the browser
- Building high-performance, client-side applications using HydroBlox
- Exploring modern web-based computing approaches for hydrology and hydroinformatics
Prerequisites:
- Programming:
- No programming experience required. Familiarity with Python and basic web technologies (HTML, JavaScript) is helpful but not required.
- Software & Data:
- All data will be accessed through the HydroBlox platform, which runs entirely in the browser.
Additional Notes:
This workshop bridges traditional hydrology and modern web technologies by focusing on hydroinformatics through client-side computing. It demonstrates how to run high-performance simulations using WebAssembly without server overhead and showcases LLM-assisted workflow automation to streamline data processing. The content is especially relevant for attendees interested in AI-driven environmental modeling, scalable web architecture, and reproducible scientific workflows.