Modelling flow, uncertainty, and transport processes in karst aquifers: theory and practical applications with the LuKARS model.
Lecturers: Dr.-Ing. Beatrice Richier (FAU Erlangen), M.Sc. Kübra Özdemir Çallı, and Prof. Andreas Hartmann (TU Dresden)
Duration: 3+3 hours
Course overview:
Karst aquifers are among the most important freshwater resources worldwide, yet their strong heterogeneity, dual flow behaviour, and limited data availability make them particularly challenging to understand, model, and protect. Reliable modelling tools are essential to simulate flow and transport processes, assess vulnerability to contamination, and support sustainable water management in karst regions.
This full‑day short course provides a comprehensive introduction to modelling karst systems, combining semi‑distributed hydrological modelling of flow processes and uncertainty with conceptual and advanced modelling of solute and contaminant transport in both the unsaturated and saturated zones.
In the first 3‑hour block, participants are introduced to the semi‑distributed conceptual modelling of karst hydrology using the LuKARS (Land use change modelling in KARSt systems) framework. The course covers the representation of spatial heterogeneity through hydrotopes, multiple storage components, and dual flow processes typical of karst systems. Participants will explore model behaviour using the latest version, LuKARS 3.0, implemented in Python. Particular emphasis is placed on equifinality, sensitivity analysis, and uncertainty quantification, including hands‑on applications of the Morris (Elementary Effects) method and the GLUE framework. The morning session will conclude with a short introduction about LuKARS REACT to model reactive solute transport in karst systems, a topic which will be further explored in the afternoon session.
In the second 3‑hour block, the focus shifts to solute and contaminant transport processes in karst aquifers, building directly on the hydrological concepts introduced earlier. Participants will learn how contaminants are transported, mixed, stored, and attenuated in karst systems, considering the roles of conduits, matrix, epikarst, and vadose zone processes. A systematic overview of transport modelling approaches is provided, ranging from simple lumped transport models to more advanced frameworks such as mobile–immobile models, transient storage concepts, and StorAge Selection (SAS) function approaches. The strengths, limitations, data requirements, and uncertainties of these models are discussed, along with guidance on selecting appropriate modelling strategies for research and applied problems.
Together, the two course blocks offer an integrated perspective on karst modelling, linking flow dynamics, uncertainty, and transport processes across spatial and temporal scales. The course builds on the short course presented at Eurokarst two years ago, with updated content, recent literature, and expanded coverage of uncertainty and transport modelling.
Who should attend:
- MSc and PhD students in hydrology, hydrogeology, and environmental sciences
- Researchers working on karst systems, groundwater modelling, or contaminant transport
- Practitioners, consultants, and water managers dealing with karst aquifers
- Current and prospective users of semi‑distributed hydrological and transport models
A basic background in hydrology or hydrogeology is recommended. No advanced programming skills are required.
Maximum number of participants: 30
Course materials:
- PowerPoint lecture slides
- Hands‑on tutorials and Jupyter Notebook scripts (Python‑based)
- Example model setups and datasets
- A curated list of key references on karst flow, uncertainty, and transport modelling
Requirements:
- Laptop with installed and working Anaconda (please check my running any python script, test run can also be found online).
- Download and tutorial for installation: https://docs.anaconda.com/anaconda/install/
- Install the list of required libraries. The list will be provided together with the course material and scripts.
Figure: LuKARS 3.0 hydrological conceptual model, including a user-defined number of hydrotopes and the uncertainty bands of spring discharge.