Quantifying the effect of soil mixing and transport on chemical weathering and clay formation
Understanding how sediment is produced, transported and weathered remains a key area of study within geomorphology. Much of this occurs in upland landscapes of interconnected hillslopes and fluvial systems where material is originally sourced from the bedrock underlying hillslopes and transported throughout the system. This conveyor belt like process combines the competing effects of erosion and weathering at the landscape scale. Erosion is required to separate bedrock and export it away allowing fresh material to subsequently be disintegrated. Weathering acts on these particles once in the soil column to transform parent material in to secondary weathering products. This gives a steady supply of secondary weathering products, often clay minerals, which provide a number of functions within the landscape. Namely the stabilisation and export of organic carbon from the hillslopes to deep oceanic sinks. However this stable flux is still poorly understood and quantified, particularly in terms of how it may vary with changing erosion rate. Constraining and exploring this relationship is therefore of prime interest in a rapidly changing world. This thesis is focused on the development of a new numerical model (LSDMxingModel) coupling a particle based geomorphological model with a reactive transport geochemical model in order to better understand these upland conveyor belts. Alongside this, existing field data and topographic analysis methods are used in order to test the model and see if the theoretical relationships can be observed at the landscape scale. To begin with I do this with a series of USA wide datasets encompassing erosion rate, soil pedon, climate, and elevation data and use topographic analysis techniques to link erosion rate with pedon data for the clay particle size fraction. Across five different landscapes in the continental USA I find three that this linkage can occur in (and two where it does not to demonstrate the results are not spurious). In them the clay particle size fraction has a negative correlation with erosion rate, expected since more erosion results in less time for parent material to breakdown in to clays within the soil. Combining this data allows the estimation for clay flux to be obtained in these landscapes for the first time and find that an optimum erosion rate exists to maximise clay flux. From this I move onto using the LSDMixingModel for further exploration of the effects of mixing and hillslope transience. Firstly I present the rationale and equations behind it followed by a series of analytical tests involving Cosmogenic Radionuclides (CRN). CRN are a common tool used to calculate erosion rate and provide a proxy for the amount of time spent within the soil column so are a useful proxy for the extent of weathering. I then present findings from a series of model runs exploring transient hillslope conditions before moving on to analysing three hillslopes with CRN soil data from the Feather River, California. I use DEM data and a new algorithm to help convert these hillslopes to the LSDMixingModel domain. I successfully apply the model to demonstrate the effects of transient conditions caused by a change in fluvial incision in both real and theoretical landscapes. Alongside this I show the effects of partial and full mixing on soil CRN concentrations and how they may have to be considered when calculating basinal erosion rates. Finally I incorporate the geochemical side of the LSDMixingModel (in the form of CrunchFlow) to model a series of theoretical scenarios to explore the potential effects of mixing, transience, and erosion on both simple and more complex parent mineralogy. I find that increasing soil mixing can have the effect of smearing or stalling the reaction front of primary Albite weathering into secondary Kaolinite. Likewise I confirm some of the theoretical ideas on optimum clay fluxes using a series of transient hillslope runs. Alongside this I attempt to isolate the effect of inheritance on the clay flux along with demonstrating that the clay fraction change propagate upslope from a base level change in a similar manner to the hillslope steepness and soil thickness. Finally I explore how erosion and mixing may affect the competing production of 2 types of clay mineral (Illite and Kaolinite) and how erosive conditions can keep the soil young enough for Illite to dominate the clay mineral fraction whereas otherwise it weathers out over time. This supports theoretical and field based findings on how the clay mineralogy might change with 2:1 clays (Illite) giving way to 1:1 clays (Kaolinite) over time.