Patterns of Ca/Sr and ~(87)Sr/~(86)Sr variation before and after a whole watershed CaSiO_3 addition at the Hubbard Brook Experimental Forest, USA

摘要

Forty-one metric tons of the mineral wollastonite (CaSiO_3) was applied to an 11.8 hectare watershed at the Hubbard Brook Experimental Forest (HBEF; White Mountains, New Hampshire, USA) with the goal of restoring the Ca estimated to have been depleted from the soil exchange complex by acid deposition. This experiment provided an opportunity to gain qualitative information on whole watershed hydrologic flow paths by studying the response of stream water chemistry to the addition of Ca. Because the Ca/Sr and ~(87)Sr/~(86)Sr ratios of wollastonite strongly contrast that of other Ca sources in the watershed, the wollastonite-derived Ca can be identified and its amount estimated in various ecosystem components. Stream water chemistry at the HBEF varies seasonally due to shifts in the proportion of base flow and interflow. Prior to the wollastonite application, seasonal variations in ~(87)Sr/~(86)Sr ratios indicated that ~(87)Sr/~(86)Sr was higher during base flow than interflow, due largely to greater amounts of biotite weathering along deeper flow paths. After the application, Ca/Sr and ~(87)Sr/~(86)Sr changed markedly as the high Ca/Sr and low ~(87)Sr/~(86)Sr wollastonite dissolved and mixed with stream water. The Ca addition provided information on the response times of various flow paths and ion exchange processes to Ca addition in this small upland watershed. During the first year after the addition, wollastonite applied to the near stream zone dissolved and was partially immobilized by cation exchange sites in the hyporheic zone. In the second and third years after the addition we infer that much of this Ca and Sr was subsequently desorbed from the hyporheic zone and was exported from the watershed in stream flow. In the fourth through ninth years after the addition, Ca and Sr from wollastonite that had dissolved in upland soils was transported to the stream by interflow during wet periods when the ground water table was elevated. Between years three and nine the minimum annual Ca/Sr ratio (in late summer base flow) increased, providing evidence that Ca and Sr had increasingly infiltrated to the deepest flow paths. Strong seasonal variations in Ca/Sr and ~(87)Sr/~(86)Sr ratios of stream water resulted from the wollastonite addition to upland forest soils, and these ratios have become sensitive to changing flow paths during the annual cycle. Most notably, high flow events now produce large excursions in stream geochemistry toward the high Ca/Sr and low ~(87)Sr/~(86)Sr ratios of wollastonite. Nine years after the application we estimate that ～360kg of Ca from wollastonite has been exported from the watershed in stream flow. The rate of export of Ca from wollastonite dissolution has stabilized at about 11kg of Ca per year, which accounts for ～30% of the dissolved Ca in the stream water. Given that 19 metric tons of Ca were applied to the watershed, and assuming this current rate of loss, it should take over 1000years for this added Ca to be transported from the watershed.