BACKGROUNDChemical weathering of carbonates (i.e.karstification) involves considerable uptake of atmospheric CO₂ which is converted to dissolve inorganic carbon (DIC), thereby acting as one of the important terrestrial carbon sinks. This karst-related carbon sink could contribute greatly to the global carbon budget and have the potential to be an increasing carbon sink on land. However, its stability has long been debated because CO₂ sequestered by the dissolution of carbonate could return to the atmosphere through CO₂ outgassing from groundwater-feeding surface waters, which can cause uncertainties for the estimation of the karst-related carbon sink.

OBJECTIVESIn order to better understand the processes responsible for CO₂ outgassing and the flux and influencing factors of CO₂ outgassing, and provide more insights into the stability of the karst-related carbon sink.

METHODSHydrochemical and isotopic techniques were used to monitor the change of water chemistry and carbon isotopic composition of dissolved inorganic δ¹³C_DIC along the flow path. Based on the downstream variations of hydrochemical indicators and δ¹³C_DIC, the flux and influencing factors of CO₂ outgassing along the stream were analyzed.

RESULTSFrom the spring (C1) to site C14, the stream channel (270m-long) had a steep gradient of~10°, and pH, calcite saturation index and δ¹³C_DIC of stream water increased by 0.9, 0.9 and 1.8‰, respectively, whereas CO₂ partial pressure, electrical conductivity, Ca²⁺ and DIC concentrations decreased by 85%, 34μS/cm, 0.2mmol/L and 0.7mmol/L, respectively. These observations indicated the occurrence of significant CO₂ degassing and calcium carbonate precipitation in the channel. In contrast, less downstream variations in water chemistry and δ¹³C_DIC of stream water occurred along C18-C26 segment (about 2.1km long, slope gradient < 1°) in the plain area, suggesting weak CO₂ outgassing and very limited calcite precipitation. Furthermore, the hydrochemical and isotopic compositions of stream water were likely to be affected by tributary mixing and dilution in the downstream area, and consequently the pH value of the stream and calcite saturation index decreased to some degrees, which inhibited the occurrence of CO₂ degassing.

CONCLUSIONSThe downstream variation in hydrochemical and isotopic compositions suggest that the stream CO₂ degassing is chiefly affected by topographically controlled hydrological conditions. At Changliushui, the CO₂ degassing in streams partly counteracts the atmospheric CO₂ sequestered by carbonate weathering, but causes 29% of the total amount of CO₂ sequestered in DIC of the feeding spring water to be released back to the atmosphere. For streams/rivers from low-relief areas fed by karst springs/underground rivers that have a large discharge rate, the CO₂ degassing should have limited impact on the stability of karst-related carbon sinks. In addition, the possibly enhanced "carbon pump" effects of aquatic phototrophs would make the karst-related carbon sink more stable.