Subdaily variations of atmospheric dimethylsulfide, dimethylsulfoxide, methanesulfonate, and non-sea-salt sulfate aerosols in the atmospheric boundary layer at Dumont d'Urville (coastal Antarctica) during summer

A study of atmospheric dimethylsulfide (DMS) and dimethylsulfoxide (DMSO) w as conducted on a subdaily basis during austral summer months (450 samples from mid-December 1998 to late-February 1999) at Dumont d'Urville, a coasta l Antarctic site (66 degrees 40'S, 140 degrees 01'E), In addition, subdaily aerosol samplings were analyzed for particulate methanesulfonate (MSA) and non-sea-salt sulfate (nssSO(4)(2-)). During these summer months, DMS and D MSO levels fluctuated from 34 to 2923 pptv (mean of 290 +/- 305 pptv) and f rom 0.4 to 57 pptv (mean of 3.4 +/- 4.4 pptv), respectively. Mean MSA and n on-sea-salt sulfate (nssSO(4)(2-)) mixing ratios were close to 12.5 +/- 8.2 pptv and 68.1 +/- 35.0 pptv, respectively. In two occasions characterized by stable wind conditions and intense insolation, it was possible to examin e the local photochemistry of DMS. During these events, DMSO levels tracked quite closely the solar flux and particulate MSA levels were enhanced duri ng the afternoons. Photochemical calculations reproduce quite well observed diurnal variations of DMSO when we assume an 0.8 yield of DMSO from the DM S/OH addition channel and an heterogeneous loss rate of DMSO proportional t o the OH radical concentration: 0.5x10(-10) [OH] + 5.5x10(-5) (in s(-1)). I f correct, on a 24 hour average the heterogeneous loss of DMSO is estimated to be 2 times faster than the DMSO/OH gas phase oxidation in these regions . Very low levels of DMSO were found in the aerosol phase (less than 0.01 p ptv), suggesting that an efficient oxidation of DMSO subsequently takes pla ce onto the aerosol surface. The observed increase of MSA levels which take s place quasi-immediately after the noon DMSO maximum suggests that an hete rogeneous oxidation of DMSO onto aerosols represents a more efficient pathw ay producing MSA compared to the gas phase DMSO/OH pathway. Since only a th ird of the total amount of DMSO lost can be explained by the observed enhan cement of MSA levels, further studies investigating other species including methanesulfinic acid and dimethylsulfone (DMSO2) formed during the oxidati on of DMS are here needed. When katabatic winds took place, bringing contin ental Antarctic air at the site, enrichments of DMSO relative to DMS and MS A relative to non-sea-salt sulfate levels were observed. That is in agreeme nt with the hypothesis of an accumulation of DMSO and probably of gaseous M SA in the free Antarctic troposphere in relation to a less efficient hetero geneous loss rate of DMSO.