SILICATE MINERAL WEATHERING, REACTION FRONTS, AND REACTIVE TRANSPORT MODELS

Silicate mineral dissolution and weathering in the critical zone has implications for processes including ecosystem nutrient cycling, soil formation, and groundwater quality. The orders of magnitude differences between laboratory and field dissolution rates result from factors including hydrology, proximity to chemical equilibrium, and surface area. A promising approach for quantifying the effects of such factors on differences between laboratory and field rates is to use a reactive transport code to model dissolution.

The Merced chronosequence is an ideal system for applying FLOTRAN, a reaction transport model, because of the large dataset and successful calculation of field dissolution rates. Key model input parameters were the 1) input chemistry, 2) input flow rate, 3) mineral volume, 4) mineral surface area (SA), 5) mineral dissolution rate constant, and 6) the exponent on the chemical affinity term in the reaction rate equation. Measured values were used for the mineral volume and laboratory values were used for the rate constants. Dissolution proceeded too quickly when these values were used with measured porewater chemistry, flow rates, BET SA, and an unmodified rate equation. Parameters 1, 2, 4, and 6 were changed as necessary to achieve a best-fit model solution. Variations in the input chemistry (within measured field concentrations) made an insignificant difference. The input flow rate was changed by less than a factor of 2 from the calculated flow rate. The largest changes were made to the affinity term and the SA because these terms are the most unconstrained. Changes from the measured SA were assumed to be a proxy for reactive SA. The affinity exponent was increased by about one magnitude to reflect results from near-equilibrium laboratory dissolution experiments. The biggest changes were made to the SA. SA needed to be changed by up to 3 orders of magnitude from the BET SA to achieve a best fit. A best fit for soils of different ages required different SA for each soil suggesting that the reactive SA changes with time (as has been hypothesized by others looking at dissolution over years to millennia). These results corroborate other studies that point to the importance of developing techniques to measure reactive SA (or reactivity of a mineral through time). Future simulations will consider dust influx effects.