Airborne measurements and model evaluation of dimethyl sulfide over the Amazon rainforest and Pacific Ocean

Abstract. Dimethyl sulfide (DMS) is the primary natural source of reduced sulfur to the atmosphere and, as a precursor to sulfate aerosols, indirectly influences Earth’s climate. Despite its global significance, upper tropospheric DMS measurements, particularly in deep-convective outflows, remain limited. Here we present two airborne datasets of in-situ DMS observations spanning 0.3–14 km over the Amazon rainforest and eastern Indo-Pacific Ocean, obtained during the CAFE-Brazil and CAFE-Pacific campaigns. These measurements confirm an east-west decreasing concentration gradient in the Amazonian boundary layer and reveal frequent elevated DMS (up to 56 pptv) in marine upper tropospheric convective outflows. Model-observation comparisons were performed with the EMAC (ECHAM5/MESSy2 Atmospheric Chemistry) model using several marine and terrestrial emission inventories and three different convection parametrisations. The sole available global terrestrial inventory (Spiro et al., 1992) overestimates Amazonian DMS by approximately a factor of three. Among the marine DMS climatologies, the Hulswar et al. (2022) inventory provides the closest agreement with the measurements, although further refinements remain10 necessary. Of the convection schemes tested, the Tiedtke scheme with Nordeng closure best reproduces the observed vertical DMS profile. Our results highlight the need for improved DMS emission parametrisations and show that the choice of convection scheme strongly influences the representation of the atmospheric sulfur cycle in global models. The frequent occurrence of DMS in deep convective outflow underscores the importance of further investigations into its transport, oxidation at cold temperatures, and role in new particle formation in the upper troposphere and stratosphere.