We are pleased to share our recent publication in Water Resources Research:
Development of a Dual-Domain Karst Flow Model Under Consideration of Preferential Film-Flow Dynamics and Analysis of Compartment-Specific Parameter Sensitivities
https://doi.org/10.1029/2024WR037677
This work was primarily carried out by Torsten Noffz as part of his PhD research, representing a major step forward in the process-based modeling of preferential flow in karst systems.
Karst aquifers are notoriously difficult to characterize and manage due to their strong heterogeneity and high vulnerability to contamination. In deep vadose zones, highly conductive features such as dissolution shafts and faults promote flow channeling and preferential recharge. These dynamics complicate the simulation of rapid infiltration processes, particularly in the context of flood forecasting and contamination risk assessment.
In this study, Torsten developed and implemented a novel modeling strategy by extending MODFLOW-CFPv2 to enable the simultaneous computation of diffuse matrix fluxes and preferential film-flow in the vadose zone. This dual-domain approach allows infiltration via preferential pathways to be represented in a physically consistent and distributed manner within an established modeling framework.
A comprehensive global sensitivity analysis of a synthetic karst system highlights the importance of explicitly representing vadose zone processes. While event-averaged sensitivities align with the commonly observed dominance of phreatic zone properties, time-dependent sensitivities reveal that during strong infiltration events, film-flow dynamics and their controlling parameters, particularly fracture facial area density and activation thresholds, can become highly influential.
The study demonstrates that sufficiently developed vadose zones with preferential pathways may not be adequately captured by bulk-effective approaches. The presented methodology supports more physically grounded karst modeling strategies and can be readily integrated into existing CFPv2 workflows.
We hope the study stimulates further advances in process-based karst modeling and risk assessment.
Noffz et al. (2026) WRR, Film flow model