Carbon mass-balance modelling and carbon isotope exchange processes in dynamic caves

Diverse interpretations have been made of carbon isotope time series in speleothems, reflecting multiple potential controls. Here we study the dynamics of C-13 and C-12 cycling in a particularly well-constrained site to improve our understanding of processes affecting speleothem delta C-13 values. The small, tubular Grotta di Ernesto cave (NE Italy) hosts annually-laminated speleothem archives of climatic and environmental changes. Temperature, air pressure, pCO(2), dissolved inorganic carbon (DIC) and their C isotopic compositions were monitored for up to five years in soil water and gas, cave dripwater and cave air. Mass-balance models were constructed for CO2 concentrations and tested against the carbon isotope data. Air advection forces winter pCO(2) to drop in the cave air to ca. 500 ppm from a summer peak of ca. 1500 ppm, with a rate of air exchange between cave and free atmosphere of approximately 0.4 days. The process of cave ventilation forces degassing of CO2 from the dripwater, prior to any calcite precipitation onto the stalagmites. This phase of degassing causes kinetic isotope fractionation, i.e. C-13-enrichment of dripwater whose delta C-13(DIC) values are already higher (by about 1 parts per thousand) than those of soil water due to dissolution of the carbonate rock. A subsequent systematic shift to even higher delta C-13 values, from -11.5 parts per thousand in the cave drips to about -8 parts per thousand calculated for the solution film on top of stalagmites, is related to degassing on the stalagmite top and equilibration with the cave air. Mass-balance modelling of C fluxes reveals that a very small percentage of isotopically depleted cave air CO2 evolves from the first phase of dripwater degassing, and shifts the winter cave air composition toward slightly more depleted values than those calculated for equilibrium. The systematic C-13-enrichment from the soil to the stalagmites at Grotta di Ernesto is independent of drip rate, and forced by the difference in pCO(2) between cave water and cave air. This implies that speleothem delta C-13 values may not be simply interpreted either in terms of hydrology or soil processes. (C) 2010 Elsevier Ltd. All rights reserved.