A stochastic modelling approach with PHREEQC: using multiple natural tracers to track solute geochemical evolution along a flowline in a forested headwater catchment

presented by Nicolaus Van Zweel

ABOUT THE SPEAKER: Environmental engineer with 6 years international experience in consulting engineering, focusing on specialist studies including data gathering and numerical modelling, forming part of risk assessments of mining related projects. The last three years of his career has been devoted to a PhD study in computational science, employed by the Luxembourg Institute for Research and Technology (LIST) in collaboration with the University of Luxembourg. I am a strong numerical modeler and competent at conceptualising and solving complex problems.

ABOUT THE TALK: 

Abstract: Modelling of solute transport and travel time distributions using a particle based Lagrangian model with a multi-tracer approach

The common notion that surface water more closely resemble soil- or ground water than precipitation has led many authors (Langbein and Dawdy, 1964; Johnson et al., 1969; Buttle and Peters, 1997; Evans and Davies, 1998) to theorize that inference can be drawn about the magnitude and mobility of stored subsurface water by observing stream chemistry (Brooks et al., 2015). It is intuitive to assume that the extent of chemical evolution of water entering the subsurface will be a function of the contact time the solute has with the host rock and soil, it is therefore fair to assume that a link can be made between stream chemistry and residence time of water in the subsurface. This task however has proven to be difficult and is one of the two central themes identified in the double paradox defined by Kirchner (2003). The first part of the paradox askes the question how ‘old water’ is so rapidly mobilized, and the second part linked to this, is the spatial variability of the water chemistry of the water stored in the catchment. The question of the first part of the double paradox has been address in several publications including in the work by KcDonnell and Beven (2014). The second part of the paradox is where integrating geochemistry into hydrological studies can add value.

One of the main pillars of this PhD project consists of the combination of quantitative (hydrological modelling) and qualitative (geochemical dataset) information. The overarching aim is to shed light on the interactions between event and pre-event water by considering, water source and mixing, and associated travel times in the subsurface. This PhD is embedded in the PRIDE DTU HYDRO-CSI (coordinated by Prof. Laurent Pfister). The collaboration between UdL, KIT (Prof. Erwin Zehe) and LIST (Dr. Christophe Hissler) via the DTU HYDRO-CSI.

The project leverages a large hydrochemical database that consist of bi-weekly observations spanning over 10 years. Traditional graphical interpretation tools such as Piper diagrams (Piper, 1944), Stiff diagrams and scatter plots are used in conjunction with more recent approaches, such as hierarchical cluster analysis (HCA) and principal component analysis (PCA) to form a conceptual geochemical model. To test hypotheses formed from the conceptual geochemical model, a combination of physical -and modelling (PHREEQC) experiments where performed.

The seminar will briefly give a background on some of the main findings of the project and will go into a bit more detail about how the numerical geochemical model was setup in PHREEQC (RStudio).

KEYWORDS: Geochemical modelling, reactive transport modelling, travel time distribution, geochemical tracers.

REGISTER ONLINE - zoom meeting