Experimental and Numerical Modelling of CO2 Atmospheric Dispersion in Hazardous Gas Emission Sites.

Abstract:

Under stable atmospheric conditions and/or in presence of topographic depressions, CO₂ concentrations can reach high values resulting in lethal effect to living organisms. The distribution of denser than air gases released from the underground is governed by gravity, turbulence and dispersion. Once emitted, the gas distribution is initially driven by buoyancy and a gas cloud accumulates on the ground (gravitational phase); with time the density gradient becomes less important due to dispersion or mixing and gas distribution is mainly governed by wind and atmospheric turbulence (passive dispersion phase). Natural analogues provide evidences of the impact of CO₂ leakage. Dangerous CO₂ concentration in atmosphere related to underground emission have been occasionally reported although the conditions favouring the persistence of such a concentration are barely studied.In this work, the dynamics of CO₂ in the atmosphere after ground emission is assessed to quantify their potential risk. Two approaches have been followed: (1) direct measurement of air concentration in a natural emission site, where formation of a “CO₂ lake” is common and (2) numerical atmospheric modelling. Two sites with different morphology were studied: (a) the Cañada Real site, a flat terrain in the Volcanic Field of Campo de Calatrava (Spain); (b) the Solforata di Pomezia site, a rough terrain in the Alban Hills Volcanic Region (Italy). The comparison between field data and model calculations reveal that numerical dispersion models are capable of predicting the formation of CO₂ accumulation over the ground as a consequence of underground gas emission. Therefore, atmospheric modelling could be included as a valuable methodology in the risk assessment of leakage in natural degassing systems and in CCS projects. Conclusions from this work provide clues on whether leakage may be a real risk for humans and under which conditions this risk needs to be included in the risk assessment.