Dust, impure calcite, and phytoliths: Modeled alternative sources of chemical weathering solutes in shallow groundwater

In highly reactive, carbonate terrains that constitute more than one-fifth of critical zone landscapes, quantifying bedrock weathering processes may require understanding the realities of carbonate mineral impurity on solubility, biotically-produced minerals as an integral part of leaky biogeochemical cycles, and dust-deposited minerals as important high-surface-area, first-contact solute sources. The potential impact of these processes has not been thoroughly investigated as groundwater solutes are mostly thought to be sourced from chemical reactions with soil and bedrock, although dust is often viewed as an important delivery mechanism of nutrients to other ecosystems, including those in mountain and tropical settings. We present results of computer and hand-calculated (manual) inverse modeling of two years of stream-water chemistry, spanning a dry and a wet year, for a groundwater-fed headwater stream at the Konza Tallgrass Prairie Long-Term Ecological Research Site in northeastern Kansas. Weathering of the limestone and shale bedrock at the site provide a possible source of solutes to the groundwater-fed stream, but our modeling suggests an alternate source considering the pathway of groundwater recharge likely encounters highly reactive phases before encountering bedrock. We used the mineralogy and geochemistry of local dust collected previously, estimates of a possible chemical composition of phytolith containing potassium, and an impure calcite representing measured limestone bedrock chemistry to show that chemical reactions with the current dust flux are adequate to account for the groundwater chemistry. Average annual amounts were about: 1) 380 kg ha(-1) yr(-1) dust dissolved, 2) 80 kg ha(-1) yr(-1) bedrock carbonate dissolved, and 3) 320 kg ha(-1) yr(-1) phytoliths precipitated. Small amounts of cation exchange were also required to balance the models. There were only small differences between the computer and manual inverse models; both methods resulted in up to 4.6 times more mass of dust dissolved than bedrock. We suggest that dust weathering may be a process that occurs widely, considering the ubiquitous dust flux in continental regions. Significance of research: Solutes in groundwater and in streams fed by groundwater are thought to be mostly derived from aquifer bedrock and soil water recharging the aquifer. This paradigm may fail for landscapes that have been subaerially exposed for 10's to 100's of millions of years in terrains where relatively rapid dissolution of minerals dominates, such as carbonate-mineral dominated rocks like limestone. Here we test whether dust, having high surface area and being the first inorganic solid encountered by meteoric precipitation, could provide enough solutes water to reduce the amount of bedrock that dissolves, extending the timeframe of landscape evolution. We found, through inverse modeling with PHREEQC, that dust can provide 4.6 times as much mass to stream water in merokarst than calcite and that the mass of dust dissolved was the same order of magnitude as local dust deposition. These results give permissive evidence that dust is important to understanding groundwater chemistry.