Saberian et al. (2025) HydroQuantum: A new quantum-driven Python package for hydrological simulation

Identification

• Journal: Environmental Modelling & Software
• Year: 2025
• Date: 2025-10-09
• Authors: Mostafa Saberian, Nima Zafarmomen, Adarsha Neupane, Krishna Panthi, Vidya Samadi
• DOI: 10.1016/j.envsoft.2025.106736

Research Groups

• The Glenn Department of Civil Engineering, Clemson University
• Department of Agricultural Sciences, Clemson University
• School of Computing, Clemson University
• Artificial Intelligence Research Institute for Science and Engineering (AIRISE), School of Computing, Clemson University

Short Summary

This research introduces "HydroQuantum," a new Python package leveraging quantum computing for hydrological simulations. It demonstrates the potential of quantum algorithms, specifically QLSTM, for daily streamflow and stream water temperature simulations, showing promising results for streamflow but underperformance for stream water temperature compared to classical LSTM.

Objective

• To leverage quantum computing for hydrological simulation by developing a new Python package, "HydroQuantum," and evaluating its performance for daily streamflow and stream water temperature simulations across the continental United States.

Study Configuration

• Spatial Scale: Continental United States
• Temporal Scale: Daily simulations; training period 2000–2014, testing period 2015–2022.

Methodology and Data

• Models used: Variational Quantum Circuits (VQC), fully quantum Long Short-Term Memory network (QLSTM), hybrid quantum-classical LSTM, classical LSTM (for benchmarking).
• Data sources: Daily streamflow and stream water temperature data (implied observational data).

Main Results

• The "HydroQuantum" Python package was successfully developed to facilitate quantum-driven hydrological simulations.
• QLSTM demonstrated impressive capabilities in capturing temporal dependencies for daily streamflow data.
• QLSTM consistently underperformed classical LSTM when simulating daily stream water temperature (SWT).
• Sensitivity analysis indicated that precipitation and snow-water equivalent were significant contributors to the performance of quantum-driven simulations.

Contributions

• Development of "HydroQuantum," a novel Python package that enables researchers to explore quantum algorithms for hydrological simulation.
• First-time implementation and benchmarking of Variational Quantum Circuits (VQC), fully quantum LSTM (QLSTM), and hybrid quantum-classical LSTM against classical LSTM for daily streamflow and stream water temperature simulations.
• Identification of key hydrological variables (precipitation and snow-water equivalent) that significantly influence quantum-driven hydrological model performance.

Funding

• Not specified in the provided text.

Citation

@article{Saberian2025HydroQuantum,
  author = {Saberian, Mostafa and Zafarmomen, Nima and Neupane, Adarsha and Panthi, Krishna and Samadi, Vidya},
  title = {HydroQuantum: A new quantum-driven Python package for hydrological simulation},
  journal = {Environmental Modelling & Software},
  year = {2025},
  doi = {10.1016/j.envsoft.2025.106736},
  url = {https://doi.org/10.1016/j.envsoft.2025.106736}
}