BOAR documentation
We present Bayesian Optimization for Automated Roughness calibration in two-dimensional hydrodynamic models (BOAR), an open-source Python framework for automated hydraulic roughness calibration in two-dimensional shallow-water models using BASEMENT. The tool combines Bayesian optimization with conditional sampling to incorporate expert knowledge, such as feasible parameter ranges and inter-parameter constraints, reducing the number of costly model evaluations. BOAR standardizes calibration workflows, improves reproducibility, and supports flexible user-defined loss functions.
How to Cite
If you use BOAR in your research, please cite:
de Oliveira, L.E.D., Franca, M.J., Huber, N.P., Vanzo, D., 2026. BOAR: Bayesian optimization for automated roughness calibration in two-dimensional hydrodynamic models. SoftwareX 35, 102899. https://doi.org/10.1016/j.softx.2026.102899
BibTeX:
@article{DEOLIVEIRA2026102899,
title = {BOAR: Bayesian optimization for automated roughness calibration in two-dimensional hydrodynamic models},
journal = {SoftwareX},
volume = {35},
pages = {102899},
year = {2026},
issn = {2352-7110},
doi = {https://doi.org/10.1016/j.softx.2026.102899},
url = {https://www.sciencedirect.com/science/article/pii/S2352711026003900},
author = {Luiz E.D. {de Oliveira} and Mário J. Franca and Nils P. Huber and Davide Vanzo},
keywords = {Bayesian optimization, Hydraulic roughness calibration, Shallow water solvers, BASEMENT, Surrogate model, Gaussian process, Model calibration},
abstract = {BOAR is an open-source Python framework for automated hydraulic roughness calibration in two-dimensional shallow-water numerical solvers. BOAR combines Bayesian optimization with conditional sampling to integrate expert knowledge, including feasible parameter ranges and inter-parameter constraints, thereby reducing the number of model evaluations. It standardizes calibration workflows, improves reproducibility, and supports flexible, user-defined loss functions. Applied to 25 laboratory vegetated compound-channel flow cases, it converged within a median of 9 model runs for a depth-error of 0.4%--1.5% tolerance. The current implementation interfaces with BASEMENT and can be extended to other river-modeling tools, providing a flexible and extensible solution for hydraulic modeling.}
}