Reproducible Hydrological Modeling with CyberGIS-Jupyter: A Case Study on SUMMA

CyberGIS-Jupyter is a cyberGIS framework for achieving data-intensive, reproducible, and scalable geospatial analytics using Jupyter Notebook based on advanced cyberinfrastructure. As a cutting-edge hydrological modeling framework, the Structure for Unifying Multiple Modeling Alternative (SUMMA) functions as a unified approach to process-based modeling. The purpose of this research is to investigate the feasibility of coupling CyberGIS-Jupyter with SUMMA to realize reproducible hydrological modeling. CyberGIS-Jupyter is employed to systematically integrate advanced cyberinfrastructure, including two high-performance computers -- Virtual ROGER and XSEDE Comet, data management, and execution and visualization of SUMMA-based modeling. By taking advantage of CyberGIS-Jupyter, users can easily tune different parameters for a SUMMA model and submit computational jobs for executing the model on HPC resources without having to possess in-depth technical knowledge about cyberGIS or cyberinfrastructure. Computational experiments demonstrate that the integration of CyberGIS-Jupyter and SUMMA achieves a high-performance and easy-to-use implementation for reproducible SUMMA-based hydrological modeling.