Bayesian hierarchical uncertainty quantification of a helical-path tortuosity predictor for modeling hydraulic conductivity in sandstone aquifers
Hydrogeology Journal · 20.11.2026
DOI: 10.1007/s10040-026-03226-9 ↗Kurz gefasst
Fachartikel in der Zeitschrift Hydrogeology Journal, online erschienen am 20.11.2026. Verfasst von Khaled Altarawneh und 1 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Southern Cross University.
Abstract (Originalsprache)
Hydraulic conductivity and effective diffusivity in saturated sandstone aquifers depend on the square of tortuosity, so using a point estimate of tortuosity leaves its contribution to uncertainty unquantified in groundwater-flow and contaminant-transport models. A Bayesian hierarchical framework quantifies uncertainty in a published helical-path tortuosity predictor using a dataset of 51 sandstone samples spanning three formation classes: Tight German sandstones, clean Fontainebleau-analogous sandstones, and unconsolidated high-connectivity sandstones. The framework propagates uncertainty to hydrogeological quantities using the existing model and published data. Posterior predictions of tortuosity agree with the observations, with a mean absolute percentage error of approximately 3.3% and conservatively calibrated predictive intervals. Convergence was assessed using rank-normalized split $$\widehat{R}$$ R ^ , effective sample sizes, and autocorrelation times, and out-of-sample performance was assessed using Pareto-smoothed importance-sampling leave-one-out and five-fold cross-validation (held-out mean absolute percentage error near 5.1%). Fixing the class critical porosities at their published values and inferring them yield statistically indistinguishable predictions, with information-criterion differences within two paired standard errors; the comparison is a robustness check, and the published thresholds remain within the credible intervals under widened and flat priors, although the data constrain them mainly from above. The tortuosity posterior is then propagated into the formation factor, the effective diffusivity and a multiplicative hydraulic-conductivity uncertainty factor, and an ensemble of one-dimensional transport simulations converts these into 90% envelopes on contaminant plume extent and on diffusive breakthrough time through a low-permeability layer, providing uncertainty bounds for models of the sandstone classes studied.
Förderung
- Southern Cross University
Zitieren
APA
Altarawneh, K., & Zaher, H. A. (2026). Bayesian hierarchical uncertainty quantification of a helical-path tortuosity predictor for modeling hydraulic conductivity in sandstone aquifers. Hydrogeology Journal. https://doi.org/10.1007/s10040-026-03226-9
Vancouver
Altarawneh K, Zaher HA. Bayesian hierarchical uncertainty quantification of a helical-path tortuosity predictor for modeling hydraulic conductivity in sandstone aquifers. Hydrogeology Journal. 2026. doi:10.1007/s10040-026-03226-9
BibTeX
@article{Altarawneh2026,
author = {Altarawneh, Khaled and Zaher, Hana Abu},
title = {Bayesian hierarchical uncertainty quantification of a helical-path tortuosity predictor for modeling hydraulic conductivity in sandstone aquifers},
journal = {Hydrogeology Journal},
year = {2026},
doi = {10.1007/s10040-026-03226-9}
}
EUVIMED