A regionally based method to identify lithology-specific hydraulic conductivity distributions in shallow aquifers using catchment-scale effective values

We present a novel method for estimating lithology-specific hydraulic conductivities from catchment-scale effective values, each interpreted as a linear combination of the conductivities of the underlying lithologies. The method particularly applies to discharge-derived estimates of effective conductivity, where river networks integrate the influence of predominantly shallow groundwater flows from adjacent hillslopes. Lithological area fractions are used as weights to formulate an overdetermined linear system, which is solved using a modified method of moments under the assumption of lognormally distributed conductivities to account for spatial variability. The method was validated on synthetic cases and applied to 113 catchments in the Armorican Massif (France). It shows good convergence properties and achieves high predictive accuracy: in 85% of cases, the modeled conductivities fall within a 90% confidence interval of the observed values. Convergence analysis indicates that using five catchments per lithology is sufficient to reach 75% predictive success, with mean and standard deviation parameters estimated within 6% and 30%, respectively. With 80 catchments, prediction accuracy increases to 80%, and parameter estimation errors are reduced to 3% and 15%. The method is computationally efficient and generalizable, offering a powerful tool to explore lithology-specific conductivity datasets at regional scales. It may also be adapted to other variables, such as porosity, where linear averaging applies, enabling broader applications in hydrogeological modeling. • New method to retrieve lithology-specific hydraulic conductivity from catchment values. • Effective tool to expand regional lithology-specific conductivity datasets. • Method validated with synthetic data and applied to 113 Armorican Massif catchments. • The method demonstrates high predictive accuracy in hydraulic conductivity estimation. • 85% of modeled conductivities fall within a 90% confidence interval of observed values.