Prieto et al. (2026) Simple Box-Cox probabilistic models for hourly streamflow predictions
Identification
- Journal: Hydrology and earth system sciences
- Year: 2026
- Date: 2026-11-09
- Authors: Cristina Prieto, Dmitri Kavetski, Fabrizio Fenicia, James W. Kirchner, David McInerney, Mark Thyer, César Álvarez
- DOI: 10.5194/hess-30-6907-2026
Research Groups
- IHCantabria â Instituto de Hidráulica Ambiental de la Universidad de Cantabria, Santander, Spain
- Department of Environmental Systems Science, ETH Zürich, Zürich, Switzerland
- Eawag, Swiss Federal Institute of Aquatic Science and Technology, Dübendorf, Switzerland
- School of Civil Engineering and Construction Management, Adelaide University, Adelaide, SA, Australia
Short Summary
This study contributes to streamflow predictions at the hourly scale by investigating practical methods for uncertainty quantification using probabilistic models. The results favor Box-Cox transformations with power parameter values of 0â0.5.
Objective
- Investigate simple and practical approaches for incorporating uncertainty in hourly streamflow predictions.
- Recommend residual error models for practical applications by comparing several heteroscedastic and autoregressive residual error models.
Study Configuration
- Spatial Scale: Catchments from Spain, Switzerland and USA that cover humid to semi-arid conditions.
- Temporal Scale: Hourly time series of rainfall, streamflow and potential evapotranspiration are employed.
Methodology and Data
- Models used: GR4H (Génie Rural à 4 paramètres Horaires) is employed for the deterministic component in Eq. (1).
- Data sources: Hourly time series of rainfall, streamflow and potential evapotranspiration from 7 catchments.
Main Results
- The results favor Box-Cox transformations with power parameter values of 0â0.5.
- The Log transformation achieves the best statistical reliability of predictions.
- Catchment-by-catchment variability becomes more pronounced when performance is stratified to the top 5â% of flows.
Contributions
- This study provides practical guidance for hydrologists and practitioners requiring uncertainty quantification at the hourly scale.
- The results are broadly consistent with earlier work with daily data, providing a seamless aggregation from hourly to longer scales.
Funding
- Not specified in the provided text.
Citation
@article{Prieto2026Simple,
author = {Prieto, Cristina and Kavetski, Dmitri and Fenicia, Fabrizio and Kirchner, James W. and McInerney, David and Thyer, Mark and Álvarez, César},
title = {Simple Box-Cox probabilistic models for hourly streamflow predictions},
journal = {Hydrology and earth system sciences},
year = {2026},
doi = {10.5194/hess-30-6907-2026},
url = {https://doi.org/10.5194/hess-30-6907-2026}
}
Original Source: https://doi.org/10.5194/hess-30-6907-2026