Modelling the biogeochemical cycle of silicon in soils using the reactive transport code MIN3P

We investigated the biogeochemical cycling of Si in an acidic brown soil covered by a coniferous forest (Douglas fir) based on a comprehensive data set and reactive transport modelling. Both published and original data enable us to make up a conceptual model on which the development of a numerical model is based. We modified the reactive transport code MIN3P, which solves thermodynamic and kinetic reactions coupled with vadose zone flow and solute transport. Simulations were performed for a one-dimensional heterogeneous soil profile and were constrained by observed data including daily soil temperature, plant transpiration, throughfall, and dissolved Si in solutions collected beneath the organic layer. Reactive transport modelling was first used to test the validity of the hypothesis that a dynamic balance between Si uptake by plants and release by weathering controls aqueous Si-concentrations. We were able to calibrate the model quite accurately by stepwise adjustment of the relevant parameters. The capability of the model to predict Si-concentrations was good. Mass balance calculations indicate that only 40% of the biogeochemical cycle of Si is controlled by weathering and that about 60% of Si-cycling is related to biological processes (i.e. Si uptake by plants and dissolution of biogenic Si). Such a large contribution of biological processes was not anticipated considering the temperate climate regime, but may be explained by the high biomass productivity of the planted coniferous species. The large contribution of passive Si-uptake by vegetation permits the conservation of seasonal concentration variations caused by temperature-induced weathering, although the modelling suggests that the latter process was of lesser importance relative to biological Si-cycling.