Controls on dissolved inorganic carbon and δ13C in cave waters from DeSoto Caverns: Implications for speleothem δ13C assessments

Unraveling the factors controlling the carbon chemistry and transport of carbon within extant karst systems has important implications concerning the assessment of time-series δ13C records of speleothems. Here we report the results of a 3-year study of total dissolved inorganic carbon [DIC] and δ13CDIC from cave waters at DeSoto Caverns (Southeastern USA) that offer valuable insight on carbon transport and the accompanied isotope fractionations from end-member sources to speleothems.[DIC] and δ13CDIC values of cave waters range from 0.2 to 6.0 mM and 2.7 to −12.9 (‰ VPDB), respectively. [DIC] and δ13CDIC of “seasonal drips” show seasonal, albeit noisy, variability and are inversely related (δ13CDIC = −2.49[DIC] + 0.64, r2 = 0.84). A shallow pool fed by multiple drips shows a bimodal δ13CDIC distribution with an isotopically heavier mode during winter (−4‰ to −5‰ VPDB) relative to summer months (−9‰ to −10‰ VPDB). A multi-year trend of decreasing water availability during the study period is not reflected in a response of cave water carbon chemistry suggesting that rainfall amount may not be a significant controlling factor of the carbon chemistry. Coupled cave air winter ventilation/summer stagnation and varying CO2 fluxes through the soil horizon and epikarst exert the strongest influence on seasonal [DIC] and δ13CDIC variability. Measured values of high [DIC] and low δ13CDIC from cave waters collected during the summer/early fall closely approximate isotopic equilibrium conditions. Conversely, low [DIC] and high δ13CDIC values during winter/early months indicate kinetically enhanced isotopic fractionations within the cave waters. The kinetically enhanced isotopic fractionation of HCO3- partitioned between degassed CO2 and precipitated CaCO3(1000lnα[(CO2-HCO3)+(CaCO3(AR)-HCO3)]/2)(1000lnα[(CO2-HCO3)+(CaCO3(AR)-HCO3)]/2) is greater by about a factor of two (−6.7 ± 0.3‰) relative to the same isotopic fractionation under equilibrium conditions (−3.1‰).On the basis of 14C mass balance and paired 14C–U/Th measurements we estimate that on average about ∼23% of C delivered annually by the drips to the aragonite stalagmites is derived from 14C-dead dolomite cap while the remainder of ∼77% is derived from 14C-live biomass. δ13C measurements of aragonite (n = 12) sampled from the tips of active speleothems during the summer months are consistent with theoretical aragonite δ13C values calculated using the shallow pool summer/early fall data thus confirming the δ13C seasonality in both drips and coeval aragonite. δ13C values of an active stalagmite section spanning the last 200 years show a normal distribution with a mean of −7.1 ± 1.2‰ (n = 81) and a mode of −7‰ to −8‰ that are statistically indistinguishable from the annual mean and mode of all dripwaters. Thus secular time-series δ13C records of stalagmites at DeSoto Caverns with resolving power >10−1 year will likely carry the imprints of drip annual means that record climate-driven δ13C seasonal biases.