A kinetic chemistry tagging technique and its application to modelling the stable isotopic composition of atmospheric trace gases

Abstract

Abstract. Isotope composition, in many cases, holds unique information on sources, chemical modification and sinks of atmospheric trace gases. Vital to the interpretation and use of an increasing number of isotope analyses is appropriate modelling. However, the exact implementation of isotopic information is a challenge, and often studies use simplifications which limit their applicability. Here we confer a thorough isotopic extension to MECCA, a comprehensive kinetic chemistry sub-model. To this end, we devise a generic tagging technique for the kinetic chemistry mechanisms implemented as the sub-submodel MECCA-TAG. The technique constitutes a diagnostic tool that can benefit the investigation of various aspects of kinetic chemistry schemes; at the same time, the designed numerical optimisation reduces the computational effort while keeping important details unaffected. We further focus specifically on the modelling of stable isotopic composition, including the required extensions of the approach. The results of MECCA-TAG are evaluated against the reference sub-submodel MECCA-DBL, which is implicitly full-detailed, but necessarily is sub-optimal in practical applications due to its high computational demands. Furthermore, we evaluate the elaborate carbon and oxygen isotopic mechanism by simulating the multi-isotope composition of CO and other trace gases in the CAABA/MECCA box-model. The mechanism realistically simulates the oxygen isotope composition of key species resulting from the interchange with ozone and main atmospheric reservoirs, as well as the carbon isotope signature transfer. The model adequately reproduces the isotope chemistry features for CO under the limitation of the modelling domain. In particular, the mass-independently fractionated (MIF) composition of CO due to reactions of ozone with unsaturated hydrocarbons (a source effect) versus its intrinsic MIF enrichment induced in the removal reaction via oxidation by OH is assessed. As for the simulated conditions, the ozone source effect was found to be up to +1‰ in Δ17O(CO). The versatile modelling framework we employ (the Modular Earth Submodel System, MESSy) opens the way for implementation of the novel detailed isotopic chemistry treatment in the three-dimensional atmospheric-chemistry general circulation model EMAC. We therefore also present estimates of the computational gain obtained by the developed optimisations.